Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks423 from public data
- Facial expression recognition based on Local Binary Patterns: A comprehensive study2,383
This paper empirically evaluates facial representation based on statistical local features, Local Binary Patterns, for person-independent facial expression recognition, and observes that LBP features perform stably and robustly over a useful range of low resolutions of face images, and yield promising performance in compressed low-resolution video sequences captured in real-world environments.
- Harmonious Attention Network for Person Re-identification1,463
A novel Harmonious Attention CNN (HA-CNN) model is formulated for joint learning of soft pixel attention and hard regional attention along with simultaneous optimisation of feature representations, dedicated to optimise person re-id in uncontrolled (misaligned) images.
- Semantic Autoencoder for Zero-Shot Learning921
This work presents a novel solution to ZSL based on learning a Semantic AutoEncoder (SAE), which outperforms significantly the existing ZSL models with the additional benefit of lower computational cost and beats the state-of-the-art when the SAE is applied to supervised clustering problem.
- 862
- Person Re-Identification by Support Vector Ranking772
This work converts the person re-identification problem from an absolute scoring p roblem to a relative ranking problem and develops an novel Ensemble RankSVM to overcome the scalability limitation problem suffered by existing SVM-based ranking methods.
- Learning a Deep Embedding Model for Zero-Shot Learning769
This paper proposes to use the visual space as the embedding space instead of embedding into a semantic space or an intermediate space, and argues that in this space, the subsequent nearest neighbour search would suffer much less from the hubness problem and thus become more effective.
- 747
- 719
- Feature Mining for Localised Crowd Counting677
This paper presents a single regression model based approach that is able to estimate people count in spatially localised regions and is more scalable without the need for training a large number of regressors proportional to the number of local regions.
- Learning a Discriminative Null Space for Person Re-identification669
This work proposes to overcome the SSS problem in re-id distance metric learning by matching people in a discriminative null space of the training data, which has a fixed dimension, a closed-form solution and is very efficient to compute.
- Person Re-identification by Video Ranking642
A novel model to automatically select the most discriminative video fragments from noisy image sequences of people where more reliable space-time features can be extracted, whilst simultaneously to learn a video ranking function for person re-id is presented.
- Transferable Joint Attribute-Identity Deep Learning for Unsupervised Person Re-identification637
This work introduces an Transferable Joint Attribute-Identity Deep Learning (TJ-AIDL) for simultaneously learning an attribute-semantic and identity-discriminative feature representation space transferrable to any new (unseen) target domain for re-id tasks without the need for collecting new labelled training data from the target domain.
- Transductive Multi-View Zero-Shot Learning513
A novel heterogeneous multi-view hypergraph label propagation method is formulated for zero-shot learning in the transductive embedding space that rectifies the projection shift between the auxiliary and target domains, exploits the complementarity of multiple semantic representations, and significantly outperforms existing methods for both zero- shot and N-shot recognition.
- Associating Groups of People510
A novel people group representation and a group matching algorithm are proposed that addresses changes in the relative positions of people in a group and the latter deals with variations in illumination and viewpoint across camera views.
- 469
- 454
- Person Re-Identification by Deep Joint Learning of Multi-Loss Classification426
This work designs a novel CNN architecture for Jointly Learning Multi-Loss (JLML) of local and global discriminative feature optimisation subject concurrently to the same re-id labelled information.
- 423
- 422
- Unsupervised Person Re-Identification by Soft Multilabel Learning421
The reference agent learning to represent each reference person by a reference agent in a joint embedding is introduced and the unified deep model outperforms the state-of-the-art unsupervised RE-ID methods by clear margins.
- 416
- 403
- Tracking colour objects using adaptive mixture models382
The use of adaptive Gaussian mixtures to model the colour distributions of objects is described to perform robust, real-time tracking under varying illumination, viewing geometry and camera parameters.
- Imbalanced Deep Learning by Minority Class Incremental Rectification373
These experimental results demonstrate the performance advantages and model scalability of the proposed batch-wise incremental minority class rectification model over the existing state-of-the-art models for addressing the problem of imbalanced data learning.
- 349
- Person Re-identification by Attributes345
This work proposes a novel method for re-identification that learns a selection and weighting of mid-level semantic attributes to describe people, an attribute-centric, parts-based feature representation that differs from and complements existing low-level features that rely purely on bottom-up statistics for feature selection.
- 339
- Gait recognition without subject cooperation333
It is argued that selecting the most relevant gait features that are invariant to changes in gait covariate conditions is the key to develop a gait recognition system that works without subject cooperation and an Adaptive Component and Discriminant Analysis is formulated which seamlessly integrates the feature selection method with subspace analysis for robust recognition.
- Person Re-identification: What Features Are Important?315
This study shows that certain features play more important role than others under different circumstances, and proposes a novel unsupervised approach for learning a bottom-up feature importance, so features extracted from different individuals are weighted adaptively driven by their unique and inherent appearance attributes.
- 314
- 308
- 307
- Gait recognition using gait entropy image300
A novel gait representation termed as Gait Entropy Image (GEnI) is proposed, based on computing entropy, which encodes in a single image the randomness of pixel values in the silhouette images over a complete gait cycle that captures mostly motion information and is robust to covariate condition changes that affect appearance.
- 276
- A Multi-View Nonlinear Active Shape Model Using Kernel PCA276
This work introduces a multi-view nonlinear shape model utilising 2D view-dependent constraint without explicit reference to 3D structures, and adopts Kernel PCA based on Support Vector Machines.
- Beyond Tracking: Modelling Activity and Understanding Behaviour271
A Dynamically Multi-Linked Hidden Markov Model (DML-HMM) is developed based on the discovery of salient dynamic interlinks among multiple temporal processes corresponding to multiple event classes resulting in its topology being intrinsically determined by the underlying causality and temporal order among events.
- 269
- MODELLING FACIAL COLOUR AND IDENTITY WITH GAUSSIAN MIXTURES269
It is argued that modelling person-specific probability densities in a generic face space using mixture models provides a technique applicable to all four face recognition tasks.
- 259
- Unsupervised Person Re-identification by Deep Learning Tracklet Association256
A Tracklet Association Unsupervised Deep Learning (TAUDL) framework characterised by jointly learning per-camera (within-camera) tracklet association (labelling) and cross-camera tracklet correlation by maximising the discovery of most likely tracklet relationships across camera views.
- Transductive Multi-view Embedding for Zero-Shot Recognition and Annotation251
This paper proposes a novel framework, transductive multi-view embedding, that rectifies the projection shift between the auxiliary and target domains, exploits the complementarity of multiple semantic representations, achieves state-of-the-art recognition results on image and video benchmark datasets, and enables novel cross-view annotation tasks.
- 242
- 235
- Person Re-Identification by Discriminative Selection in Video Ranking231
This work presents a novel model to automatically select the most discriminative video fragments from noisy/incomplete image sequences of people from which reliable space-time and appearance features can be computed, whilst simultaneously learning a video ranking function for person ReID.
- 230
- Class Rectification Hard Mining for Imbalanced Deep Learning227
This work develops an end-to-end deep learning framework capable of avoiding the dominant effect of majority classes by discovering sparsely sampled boundaries of minority classes by introducing a Class Rectification Loss (CRL) regularising algorithm.
- 227
- Crowd Counting and Profiling: Methodology and Evaluation226
This study describes and compares the state-of-the-art methods for video imagery based crowd counting, and provides a systematic evaluation of different methods using the same protocol.
- 221
- Dynamic Vision214
Many of the issues raised are relevant to object recognition in general, and such visual learning machines have numerous potential applications in areas such as visual surveillance multimedia and visually mediated interaction.
- Visual surveillance in a dynamic and uncertain world210
It is demonstrated that using expectations of object trajectory, size and speed for the particular scene improves robustness and sensitivity in dynamic tracking and segmentation and behavioral evaluation under attentional control using a combination of a static BBN TASKNET and dynamic network.
- Person Re-identification199
This text explores the design rationale and implementation considerations of building a practical re-identification system, and introduces examples of robust feature representations and reviews salient feature weighting and selection mechanisms.
- 194
- Unsupervised Tracklet Person Re-Identification188
This work forms an Unsupervised Tracklet Association Learning (UTAL) framework that is capable of incrementally discovering and exploiting the underlying re-id discriminative information from automatically generated person tracklet data end-to-end.
- Recent Advances in Zero-Shot Recognition: Toward Data-Efficient Understanding of Visual Content186
This article provides a comprehensive review of existing zero-shot recognition techniques covering various aspects ranging from representations of models, data sets, and evaluation settings and highlights the limitations of existing approaches.
- Spectral clustering with eigenvector selection185
A novel spectral clustering algorithm is proposed which differs from previous approaches in that only informative/relevant eigenvectors are employed for determining the number of clusters and performing clustering and the key element of the proposed algorithm is a simple but effective relevance learning method.
- Support vector machine based multi-view face detection and recognition183
A novel algorithm for face detection is developed by combining the Eigenface and SVM methods which performs almost as fast as theEigenface method but with a significant improved speed.
- Semi-Supervised Learning under Class Distribution Mismatch182
This work addresses under-studied and realistic SSL problem by a novel algorithm named Uncertainty-Aware Self-Distillation (UASD), which produces soft targets that avoid catastrophic error propagation, and empower learning effectively from unconstrained unlabelled data with out-of-distribution (OOD) samples.
- Person Search by Multi-Scale Matching179
This work proposes a Cross-Level Semantic Alignment (CLSA) deep learning approach capable of learning more discriminative identity feature representations in a unified end-to-end model that favourably eliminates the need for constructing a computationally expensive image pyramid and a complex multi-branch network architecture.
- Transductive Zero-Shot Action Recognition by Word-Vector Embedding176
This study constructs a mapping between visual features and a semantic descriptor of each action category, allowing new categories to be recognised in the absence of any visual training data, and achieves the state-of-the-art zero-shot action recognition performance with a simple and efficient pipeline, and without supervised annotation of attributes.
- Multi-camera Matching using Bi-Directional Cumulative Brightness Transfer Functions176
A Cumulative Brightness Transfer Function (CBTF) is proposed for mapping colour between cameras located at different physical sites, which makes better use of the available colour information from a very sparse training set and a bi-directional mapping approach is developed to obtain a more accurate similarity measure between a pair of candidate objects.
- 176
- 175
- RGB-IR Person Re-identification by Cross-Modality Similarity Preservation172
This work designs a modality-gated node as a universal representation of both modality-specific and shared structures for constructing a structure-learnable feature extractor called Modality-Gated Extractor, and proposes a Focal Modality-Aware Similarity-Preserving Loss.
- 171
- 165
- Peer Collaborative Learning for Online Knowledge Distillation163
This work proposes a novel Peer Collaborative Learning method for online knowledge distillation, which integrates online ensembling and network collaboration into a unified framework and significantly improves the generalisation of various backbone networks and outperforms the state-of-the-art methods.
- Attribute Recognition by Joint Recurrent Learning of Context and Correlation162
The JRL model learns jointly pedestrian attribute correlations in a pedestrian image and in particular their sequential ordering dependencies (latent high-order correlation) in an end-to-end encoder/ decoder recurrent network.
- 161
- Low-Resolution Face Recognition157
There is a significant gap between the reported FR performances on popular benchmarks and the results on TinyFace, and the advantages of the proposed CSRI over a variety of state-of-the-art FR and super-resolution deep models on solving this largely ignored FR scenario.
- 156
- 145
- 139
- Colour model selection and adaptation in dynamic scenes137
This work uses colour mixture models for real-time colour-based object localisation, tracking and segmentation in dynamic scenes within a framework that addresses the issues of model order selection, modelling scene background and model adaptation in time.
- 130
- Segmentation and tracking using colour mixture models127
A system is described that provides robust and real-time focus-of-attention for tracking and segmentation of multi-coloured objects and dynamic body segmentation for virtual studios based upon combined foreground and background models.
- Dictionary Learning with Iterative Laplacian Regularisation for Unsupervised Person Re-identification126
A new dictionary learning for sparse coding formulation with a graph Laplacian regularisation term whose value is set iteratively enables the exploitation of cross-view identity-discriminative information ignored by existing unsupervised Re-ID methods.
- 125
- Beyond Facial Expressions: Learning Human Emotion from Body Gestures125
This paper investigates affective body gesture analysis in video sequences, a relatively understudied problem, and presents to fuse facial expression and body gesture at the feature level using Canonical Correlation Analysis (CCA).
- Deep Low-Resolution Person Re-Identification124
This work develops a novel approach called Super-resolution and Identity joiNt learninG (SING) to simultaneously optimise image super- resolution and person re-id matching and introduces an adaptive fusion algorithm for accommodating multi-resolution LR images.
- 121
- Multi-Task Zero-Shot Action Recognition with Prioritised Data Augmentation120
A visual-semantic mapping with better generalisation properties and a dynamic data re-weighting method to prioritise auxiliary data that are relevant to the target classes are introduced and applied to the challenging zero-shot action recognition problem.
- Person re-identification by unsupervised video matching120
This work proposes an unsupervised approach to person re-identification based on typical surveillance image-sequences, and presents a new video representation particularly tailored for person ReID, built up on existing action space-time features.
- Semantic embedding space for zero-shot action recognition120
This paper addresses zero-shot recognition in contemporary video action recognition tasks, using semantic word vector space as the common space to embed videos and category labels, and demonstrates that a simple self-training and data augmentation strategy can significantly improve the efficacy of this mapping.
- Cross View Gait Recognition Using Correlation Strength119
A novel approach to cross-view gait recognition with the view angle of a probe gait sequence unknown is developed, which can cope with feature mis-match across view and is more robust against feature noise.
- Attribute Learning for Understanding Unstructured Social Activity118
This paper contributes an unstructured social activity attribute (USAA) dataset, introduces the concept of semi-latent attribute space, and proposes a novel model for learning the latent attributes which alleviate the dependence of existing models on exact and exhaustive manual specification of the attribute-space.
- 118
- Attributes-Based Re-identification116
This chapter shows how mid-level “semantic attributes” can be computed for person description and shows how this attribute-based description can be used in synergy with low-level feature descriptions to improve re-identification accuracy when an attribute-centric distance measure is employed.
- Video Behaviour Mining Using a Dynamic Topic Model116
The MCTM builds on existing dynamic Bayesian network models and Bayesian topic models, and overcomes their drawbacks on sensitivity, robustness and efficiency, and profiles complex dynamic scenes by robustly clustering visual events into activities and these activities into global behaviours with temporal dynamics.
- Human-in-the-Loop Person Re-identification115
A novel human-in-the-loop re-id model based on Human Verification Incremental Learning (HVIL) is formulated which does not require any pre-labelled training data to learn a model, therefore readily scalable to new camera pairs, and is formulated to combine a series of incrementally learned HVIL models into a single ensemble model to be used when human feedback becomes unavailable.
- 115
- Person Re-Identification by Unsupervised $$\ell _1$$ Graph Learning112
This work proposes a novel unsupervised Re-ID approach which requires no labelled training data yet is able to capture discriminative information for cross-view identity matching, and is based on a new graph regularised dictionary learning algorithm.
- Unsupervised Learning of Generative Topic Saliency for Person Re-identification112
This document is distributed unchanged freely in print or electronic forms because the authors believe it is in the public interest to do so.
- The Re-identification Challenge110
This chapter presents an overview of recent progress and the state-of-the-art approaches to solving some of the fundamental challenges in person re-identification, benefiting from research in computer vision, pattern recognition and machine learning, and drawing insights from video analytics system design considerations for engineering practical solutions.
- Incremental and adaptive abnormal behaviour detection109
The experimental results show that the incremental and adaptive behaviour modelling approach is superior to a conventional batch-mode one in terms of both performance on abnormality detection and computational efficiency.
- Towards Person Identification and Re-identification with Attributes108
This work shows how mid-level semantic attributes can be used synergistically with low-level features for both identification and re-identification, and learns an attribute-centric representation to describe people, and a metric for comparing attribute profiles to disambiguate individuals.
- Composite support vector machines for detection of faces across views and pose estimation106
This work extends SVMs to model the appearance of human faces which undergo non-linear change across multiple views and uses inherent factors in the nature of the input images and the SVM classification algorithm to perform both multi-view face detection and pose estimation.
- 106
- Tracking facial feature points with Gabor wavelets and shape models106
A feature-based approach to tracking rigid and non-rigid facial motion is described, using a flexible shape model to impose global constraints upon the local feature points and to constrain the tracker.
- Global Behaviour Inference using Probabilistic Latent Semantic Analysis105
A semantic scene segmentation model is proposed to decompose a wide-area scene into regions where behaviours share similar characteristic and are represented as classes of video events bearing similar features to infer global behaviour patterns.
- Free-hand sketch recognition by multi-kernel feature learning103
This work proposes a Multiple Kernel Learning (MKL) framework for sketch recognition, fusing several features common to sketches, and investigates the use of attributes as a high-level feature for sketches and shows how this complements low-level features for improving recognition performance under the MKL framework.
- 101
- 100
- 99
- Deep Learning Logo Detection with Data Expansion by Synthesising Context98
This work designs a novel algorithm for generating Synthetic Context Logo (SCL) training images to increase model robustness against unknown background clutters, resulting in superior logo detection performance.
- Tracklet Self-Supervised Learning for Unsupervised Person Re-Identification97
A novel tracklet self-supervised learning (TSSL) method, which is capable of capitalising directly from abundant unlabelled tracklet data, to optimise a feature embedding space for both video and image unsupervised re-id.
- Appearance Manifold of Facial Expression96
The proposed generalized appearance manifold provides a unified framework for automatic facial expression analysis and is proposed to define a global coordinate space, and align subjects on one generalized manifold.
- 94
- 93
- Multi-task Curriculum Transfer Deep Learning of Clothing Attributes92
A novel Multi-Task Curriculum Transfer (MTCT) deep learning method to explore multiple sources of different types of web annotations with multi-labelled fine-grained attributes for model transfer learning from well-controlled shop clothing images collected from web retailers to in-the-wild images from the street.
- Interestingness Prediction by Robust Learning to Rank91
A more principled way to identify annotation outliers is proposed by formulating theinterestingness prediction task as a unified robust learning to rank problem, tackling both the outlier detection and interestingness prediction tasks jointly.
- Fusing appearance and distribution information of interest points for action recognition90
This paper proposes a novel action representation method which differs significantly from the existing interest point based representation in that only the global distribution information of interest points is exploited and holistic features from clouds ofinterest points accumulated over multiple temporal scales are extracted.
- 89
- Face distributions in similarity space under varying head pose89
The results show that orientation-selective Gabor filters enhance differences in pose and that different filter orientations are optimal at different poses, while principal component analysis was found to provide an identity-invariant representation in which similarities can be calculated more robustly.
- Cross-Sentence Temporal and Semantic Relations in Video Activity Localisation88
This work develops a more accurate weakly-supervised solution by introducing Cross-Sentence Relations Mining (CRM) in video moment proposal generation and matching when only a paragraph description of activities without per-sentence temporal annotation is available.
- Head Pose Classification in Crowded Scenes88
A new head pose descriptor is formulated using similarity distance maps by indexing each pixel of a head image to the mean appearance templates of head images at different poses, which are then used to train a multi-class Support Vector Machine for pose classification.
- Gait Representation Using Flow Fields88
This paper proposes a novel gait representation based on optical flow fields computed from normalized and centred person images over a complete gait cycle that is not only more discriminative, but also less sensitive to changes in various covariate conditions including clothing, carrying, shoe, and speed.
- 88
- 86
- On-the-fly feature importance mining for person re-identification86
This study proposes a novel unsupervised approach to bottom-up feature importance mining on-the-fly specific to each re-identification probe target image, so features extracted from different individuals are weighted adaptively driven by their salient and inherent appearance attributes.
- 85
- Scene Segmentation for Behaviour Correlation85
A novel framework for detecting abnormal pedestrian and vehicle behaviour by modelling cross-correlation among different co-occurring objects both locally and globally in a given scene is presented.
- Action categorization with modified hidden conditional random field82
Effective silhouette-based action features are extracted using motion moments and spectrum of chain code in a modified HCRF to have a guaranteed global optimum in the modelling of the temporal action dependencies after the HMM pathing stage.
- 82
- Tracking Multiple People Under Occlusion Using MultipleCameras82
A system for tracking multiple people with multiple cameras based on fusion of multiple cues and results demonstrate that the system can deal with the occlusion.
- Object tracking using adaptive colour mixture models82
The use of adaptive Gaussian mixtures to model the colour distributions of objects is described to perform robust, real-time tracking under varying illumination, viewing geometry and camera parameters.
- Detecting and discriminating behavioural anomalies80
The proposed framework outperforms existing methods in detecting durational anomalies in complex behaviours and subtle anomalies that are difficult to detect when objects are viewed in isolation, allowing for a more effective detection of subtle anomalies in surveillance videos.
- Sketch Recognition by Ensemble Matching of Structured Features78
This work presents a method for the representation and matching of sketches by exploiting not only local features but also global structures of sketches, through a star graph based ensemble matching strategy, and shows that by encapsulating holistic structure matching and learned bag-of-features models into a single framework, notable recognition performance improvement can be observed.
- 73
- Combining global, regional and contextual features for automatic image annotation73
A novel approach to automatic image annotation which combines global, regional, and contextual features by an extended cross-media relevance model that describes the global features as a distribution vector of visual topics and model the textual context as a multinomial distribution.
- Security and Surveillance72
A significant application of video analysis and understanding is intelligent surveillance, which aims to interpret automatically human activity and detect unusual events that could pose a threat to public security and safety.
- Fusing gait and face cues for human gender recognition72
This paper investigates gender classification from human gaits in image sequences, and proposes to fuse gait and face for improved gender discrimination using canonical correlation analysis (CCA), a powerful tool that is well suited for relating two sets of measurements.
- Re-id: Hunting Attributes in the Wild71
This document is intended to assist in the preparation of future generations of interpreters and interpreters for the deaf and hard of hearing and is suitable for use with interpreters.
- Face recognition in dynamic scenes71
An integrated system for the acquisition, normalisation and recognition of moving faces in dynamic scenes using mixture models and the use of Gaussian colour mixtures for face detection and tracking is introduced.
- Semi-supervised Deep Learning with Memory70
This work proposes a novel Memory-Assessment Deep Neural Network capable of exploiting the memory of model learning to enable semi-supervised learning and introduces a memory mechanism into the network training process as an assimilation-accommodation interaction between the network and an external memory module.
- Analysis and Modelling of Faces and Gestures70
A Dynamic Classification Method to Detect Faces and Identify Facial Expression and a Binary Decision Tree Implementation of a Boosted Strong Classifier for Face Recognition.
- Generalising without Forgetting for Lifelong Person Re-Identification69
This work forms a new Generalising without Forgetting method (GwFReID) for lifelong Re-ID and design a comprehensive learning objective that accounts for classification coherence, distribution coherence and representation coherence in a unified framework, which enables GwF re-identification to incrementally improve model generalisation without catastrophic forgetting.
- Ranking Distance Calibration for Cross-Domain Few-Shot Learning68
This work starts from a generic representation pre-trained by a cross-entropy loss and a conventional distance-based classifier, along with an image retrieval view, to employ a re-ranking process to calibrate a target distance matrix by discovering the k-reciprocal neighbours within the task.
- 68
- Discriminative Topics Modelling for Action Feature Selection and Recognition68
A novel framework for recognising realistic human actions in unconstrained environments based on computing a rich set of descriptors from key point trajectories is presented and an adaptive feature fusion method to combine different local motion descriptors for improving model robustness against feature noise and background clutters is developed.
- Face Tracking and Pose Representation68
A dynamic face tracking system based on an integrated motion-based object tracking and model-based face detection framework that produces segmented face sequences from complex scenes with poor viewing conditions in surveillance applications is described.
- Vehicle Re-identification in Context67
Extensive evaluations show that the proposed multi-scale method outperforms the state-of-the-art vehicle re-id methods on three benchmark datasets: VehicleID, VeRi-776, and VRIC (Available at http://qmul-vric.io).
- Deep Reinforcement Learning Attention Selection For Person Re-Identification67
This work develops a joint learning deep model that optimises person re-id attention selection within any auto-detected person bounding boxes by reinforcement learning of background clutter minimisation subject to re-ID label pairwise constraints.
- Dynamic Facial Expression Recognition Using A Bayesian Temporal Manifold Model67
A novel Bayesian approach to modelling temporal transitions of facial expressions represented in a manifold resulting in both superior recognition rates and improved robustness against static frame-based recognition methods is proposed.
- 66
- 64
- Fine-grained sketch-based image retrieval by matching deformable part models63
This document is distributed unchanged freely in print or electronic forms because the authors believe it to be of high quality and fit for purpose of publication.
- Faster Person Re-identification62
This work designs an All-in-One (AiO) framework together with a Distance Threshold Optimization (DTO) algorithm, which simultaneously learn and enhance multiple codes of different lengths in a single model, which complementarily uses short and long codes, achieving both faster speed and better accuracy.
- 62
- 60
- Recognition of human gestures and behaviour based on motion trajectories60
This paper introduces a statistical dynamic framework to model and recognise human activities based on learning prior and continuous propagation of density models of behaviour patterns by addressing problems of probabilistic and uncertain nature of motion patterns, non-linear temporal scaling and ambiguities in temporal segmentation.
- 59
- Fusion of perceptual cues for robust tracking of head pose and position59
A framework for fusing di!erent information sources through estimation of covariance from observations is proposed and demonstrated in a face and 3D pose tracking system that fuses similarity-to-prototypes measures and skin colour to track head pose and face position simultaneously.
- 58
- Recognising trajectories of facial identities using kernel discriminant analysis58
Kernel discriminant analysis, which employs the kernel technique to perform linear discriminantAnalysis in a high-dimensional feature space, is developed to extract the significant nonlinear features which maximise the between- class variance and minimise the within-class variance.
- Free-Hand Sketch Synthesis with Deformable Stroke Models57
A generative model which can automatically summarize the stroke composition of free-hand sketches of a given category is presented, which represents both consistent as well as diverse aspects of each sketch category.
- 57
- 56
- Transductive Multi-label Zero-shot Learning56
This paper proposes a multi-output deep regression model to project an image into a semantic word space, which explicitly exploits the correlations in the intermediate semantic layer of word vectors, and a novel zero-shot learning algorithm for multi-label data that exploits the unique compositionality property of semantic word vector representations.
- TC-Net for iSBIR55
A Triplet Classification Network (TC-Net) for iSBIR is presented which is composed of two major components: triplet Siamese network, and auxiliary classification loss which can break the limitations existed in previous works.
- 55
- Robust Subjective Visual Property Prediction from Crowdsourced Pairwise Labels55
A more principled way to identify annotation outliers is proposed by formulating the subjective visual property prediction task as a unified robust learning toRank problem, tackling both the outlier detection and learning to rank jointly.
- Activity based surveillance video content modelling55
To segment a video based on activity, a semantically meaningful video content representation method is proposed and two segmentation algorithms are proposed, one being offline offering high accuracy in segmentation, and the other being online enabling real-time performance.
- 54
- Conditional Mutual Infomation Based Boosting for Facial Expression Recognition54
Extensive experiments on the Cohn-Kanade database illustrated that LBP features are effective for expression analysis, and CMIB enables much faster training than AdaBoost, and yields a classifier of improved c lassification performance.
- Towards Generalisable Video Moment Retrieval: Visual-Dynamic Injection to Image-Text Pre-Training52
This work explores multi-modal correlations derived from large-scale image-text data to facilitate generalisable VMR and proposes a generic method, referred to as Visual-Dynamic Injection (VDI), to empower the model's understanding of video moments.
- Learning Behavioural Context52
A novel framework for automatic discovering and learning of behavioural context for video-based complex behaviour recognition and anomaly detection and a method for detecting subtle behavioural anomalies against the learned context are developed.
- Multi-camera Matching under Illumination Change Over Time51
This paper proposes a novel method for multi-camera object association based on adapting a learned inter- camera illumination mapping function to new illumination conditions over time without the need for a manual training stage using new fore- ground objects.
- 51
- 49
- Global Abnormal Behaviour Detection Using a Network of CCTV Cameras49
A novel framework is proposed, which consists of object tracking across camera views, global behaviour modelling based on unsupervised learning, and probabilistic abnormality inference, and the effectiveness of the framework is demonstrated with experiments on real-world surveillance video data.
- 48
- Action categorization by structural probabilistic latent semantic analysis48
This work proposes a new approach structural pLSA (SpLSA) to model explicitly word orders by introducing latent variables, and develops an action categorization approach that learns action representations as the distribution of latent topics in an unsupervised way.
- 47
- Autonomous Visual Events Detection and Classification without Explicit Object-Centred Segmentation and Tracking47
A new approach for modelling such temporal events based on the local intensity temporal history of pixels is proposed, which provides a computationally very effective temporal measure for detecting autonomous events.
- Intra-Camera Supervised Person Re-Identification46
MATE is designed for self-discovering the cross-camera identity correspondence in a per-camera multi-task inference framework and significantly outperforming unsupervised learning models and closely approaching conventional fully supervised learning competitors.
- 46
- Stream-Based Active Unusual Event Detection46
Experimental results on busy public space videos show that with minimal human supervision, the approach outperforms existing supervised and unsupervised learning strategies in identifying unusual events.
- 46
- Non-intrusive person authentication for access control by visual tracking and face recognition46
Work aimed at performing face recognition in more unconstrained environments such as occur in security applications based on closed-circuit television (CCTV) and an application in non-intrusive access control is discussed.
- 45
- Constructing Facial Identity Surfaces for Recognition45
A novel approach to face recognition is presented by constructing facial identity structures across views and over time, referred to as identity surfaces, in a Kernel Discriminant Analysis (KDA) feature space, aimed at addressing three challenging problems in face recognition.
- 45
- Multi-Task Mutual Learning for Vehicle Re-Identification44
A novel Multi-Task Mutual Learning (MTML) deep model to learn discriminative features simultaneously from multiple branches to de-sign a consensus learning loss function by fusing features from the convolutional feature maps from all branches.
- Investigating Open-World Person Re-identification Using a Drone44
This paper introduces the first dataset for mobile re-identification, and uses this to elucidate the unique challenges of mobile re,id models and suggest future avenues for research in this area.
- 44
- Sparse Multiscale Local Binary Patterns43
A pairwise-coupled reformulation of LBP-type classificatio n which involves selecting single-point features for each pair of classes to form compact, contextually-relevant multiscale predicates known as Multiscale Selected Local Binary Features (MSLBF).
- 43
- Single-Label Multi-Class Image Classification by Deep Logistic Regression42
This work develops two novel LR based objective functions that not only generalise the conventional LR but importantly turn out to be competitive alternatives to SR in single label classification.
- 41
- Decentralised Learning from Independent Multi-Domain Labels for Person Re-Identification40
This work proposes a novel paradigm called Federated Person Re-Identification (FedReID) to construct a generalisable global model (a central server) by simultaneously learning with multiple privacy-preserved local models (local clients) and achieves compelling generalisation performance beyond any locally trained models without using shared training data.
- 40
- 40
- Tracking Discontinuous Motion Using Bayesian Inference40
This work presents a method for tracking the head and hands of a human subject from a single view with no constraints on the continuity of motion, appropriate for real-time applications in which the availability of visual data is constrained, and motion is discontinuous.
- Face re-identification challenge: Are face recognition models good enough?39
A true large scale Surveillance Face Re-ID benchmark (SurvFace) is introduced, characterised by natively low-resolution, motion blur, uncontrolled poses, varying occlusion, poor illumination, and background clutters, where facial images are captured and detected under realistic surveillance scenarios.
- Intra-Camera Supervised Person Re-Identification: A New Benchmark38
The Multi-Task Multi-Label (MTML) deep learning method is specially designed for self-discovering the inter-camera identity correspondence and can also efficiently learn the discriminative re-id feature representations by fully using the available identity labels within each camera-view.
- 38
- Refining image annotation using contextual relations between words38
A probabilistic approach to refine image annotations by incorporating semantic relations between annotation words using a conditional random field (CRF) model where each vertex indicates the final decision on a candidate annotation word.
- Scalable Person Re-Identification by Harmonious Attention36
This work forms a novel harmonious attention network (HAN) framework to jointly learn soft pixel attention and hard region attention alongside simultaneous deep feature representation learning, particularly enabling more discriminative re-id matching by efficient networks with more scalable model inference and feature matching.
- Modelling Multi-object Activity by Gaussian Processes36
Experimental results on a public traffic scene show that the GP models outperform DBNs in terms of anomaly sensitivity, noise robustness, and flexibility in modelling complex activity.
- 36
- Enhancing Zero-Shot Facial Expression Recognition by LLM Knowledge Transfer35
A novel method is proposed, Exp-CLIP, to enhance zero-shot FER by transferring the task knowledge from large language models (LLMs), which achieves superior zero-shot results to the CLIP models and several other large vision-language models (LVLMs) on seven in-the-wild FER datasets.
- Training-free Zero-shot Composed Image Retrieval with Local Concept Reranking35
This work presents a new training-free zero-shot composed image retrieval method which translates the query into explicit human-understandable text, which helps improve model learning efficiency to enhance the generalization capacity of foundation models.
- Video Analytics for Business Intelligence35
This book presents the latest developments on video analytics for business intelligence applications and provides both academic and commercial practitioners an understanding of the state-of-the-art and a resource for potential applications and successful practice.
- Open-world Person Re-Identification by Multi-Label Assignment Inference34
This document is distributed unchanged freely in print or electronic forms because the authors believe it is in the public interest to do so.
- On the semantics of visual behaviour, structured events and trajectories of human action34
The problem of modelling the semantics of visual events without segmentation or computation of object-centred trajectories is addressed and the effectiveness of spatio-temporal dynamics based prediction to that of reasoning about body-part associations based on modelling semantics using Bayesian belief networks is compared.
- 34
- Learning to Associate Faces across Views in Vector Space of Similarities to Prototypes34
An automatic face data acquisition system based on a magnetic sensor and a calibrated camera enabled us to obtain systematically a database of face images with labelled 3D poses across a view-sphere of yaw and tilt at intervals of .
- Person Re-identification in Identity Regression Space33
This work proposes to construct an identity regression space (IRS) based on embedding different training person identities (classes) and formulate re-id as a regression problem solved by identity regression in the IRS, characterised by a closed-form solution with high learning efficiency and an inherent incremental learning capability with human-in theloop.
- 33
- Tracking and Recognition of Face Sequences33
This work detects and track a moving head before segmenting face images from on-line camera inputs and measures temporal changes in the pattern vectors of eigenface projections of successive image frames of a face sequence and introduces the concept of “temporal signature” of aFace class.
- Neural Graph Embedding for Neural Architecture Search32
This work introduces a novel idea of neural graph embedding (NGE), which represents the building block of neural networks with a neural DAG, and learns it by leveraging a Graph Convolutional Network to propagate and model the intrinsic topology information of network architectures.
- 32
- 32
- 32
- Learning pixel-wise signal energy for understanding semantics32
A new pixel-level approach for learning the temporal characteristics of change at individual pixels is proposed, using Gaussian mixture models to model slow long-term changes in pixel distributions and pixel energy histories to extract fast-change signatures from short-term events.
- CoS: Chain-of-Shot Prompting for Long Video Understanding31
Chain-of-Shot prompting is proposed, to frame shot selection as test-time visual prompt optimisation, choosing shots adaptive to video understanding semantic task by optimising shots-task alignment.
- 31
- Joint Bilateral-Resolution Identity Modeling for Cross-Resolution Person Re-Identification30
An important discovery is that leveraging LR-specific identity information enables a simple cascade of super-resolution and person re-id learning to achieve state-of-the-art performance, without elaborate model design nor bells and whistles, which has not been investigated before.
- Domain transfer for person re-identification30
This paper shows how to leverage prior re-identification models learned for a set of source view pairs, and flexibly combine these to obtain good re-Identification performance in a target view pair with greatly reduced training data requirements in the target domain.
- Bayesian Modality Fusion for Tracking Multiple People with a Multi-Camera System30
A multi-camera system based on Bayesian modality fusion to track multiple people in an indoor environment that can maintain people’s identities by using multiple cameras cooperatively is presented.
- 30
- Zero-Shot Video Moment Retrieval from Frozen Vision-Language Models29
This work proposes a zero-shot method for adapting generalisable visual-textual priors from arbitrary VLM to facilitate moment-text alignment, without the need for accessing the VMR data, and devise a conditional feature refinement module to generate boundary-aware visual features conditioned on text queries to enable better moment boundary understanding.
- Unsupervised cross-domain person re-identification by instance and distribution alignment29
A novel Hierarchical Un-supervised Domain Adaptation (HUDA) method that can transfer labelled information of an existing dataset (a source domain) to an unlabelled target domain for unsupervised person re-id and overcome the under-constrained learning problem of existing unsuper supervised domain adaptation methods.
- 29
- Generating Structured Pseudo Labels for Noise-resistant Zero-shot Video Sentence Localization28
A Structure-based Pseudo Label generation (SPL) is proposed, which first generate free-form interpretable pseudo queries before constructing query-dependent event proposals by modeling the event temporal structure.
- 28
- On Utilising Template and Feature-based Correspondence in Multi-view Appearance Models28
This paper utilises a generic-view template, a generic 3D surface model and Kernel PCA for modelling shape and texture nonlinearities across views and examines the use of affine and local feature-based transformations in establishing correspondences between very large pose variations.
- Scalable logo detection by self co-learning27
This work proposes a novel incremental learning approach, called Scalable Logo Self-co-Learning (SL2), capable of automatically self-discovering informative training images from noisy web data for progressively improving model capability in a cross-model co-learning manner.
- Self-Referenced Deep Learning27
Self-Referenced Deep Learning improves the model generalisation performance compared to vanilla learning and conventional knowledge distillation approaches with negligible extra computational cost and enhanced deployment performance on both coarse-grained object categorisation tasks.
- 27
- Highly Efficient Regression for Scalable Person Re-Identification27
A Highly Efficient Regression (HER) model is formulated by embedding the Fisher's criterion to a ridge regression model for very fast re-id model learning with scalable memory/storage usage and shows that such a simple and fast model not only outperforms notably the state-of-the-art re-ID methods, but also is more scalable to large data with additional benefits to active learning.
- 27
- Learning from Multiple Sources for Video Summarisation26
This paper presents a novel unsupervised framework to learn jointly from both visual and independently-drawn non-visual data sources for discovering meaningful latent structure of surveillance video data, and investigates ways to cope with discrepant dimension and representation whilst associating these heterogeneous data sources.
- 26
- 25
- 25
- 25
- ViMo: A Generative Visual GUI World Model for App Agents24
This work proposes ViMo, the first visual world model designed to generate future App observations as images and proposes a novel data representation, the Symbolic Text Representation~(STR) to overlay text content with symbolic placeholders while preserving graphics.
- 3D Shape Temporal Aggregation for Video-Based Clothing-Change Person Re-identification24
A 3D Shape Temporal Aggregation (3STA) model for video-based clothing-change Re-ID is proposed and an identity-aware 3D shape generation module is introduced that embeds the identity information into the generation of 3D shapes by the joint learning of shape estimation and identity recognition.
- Hierarchical distillation learning for scalable person search24
This work investigates the scalability problem of person search involving both model accuracy and inference efficiency simultaneously, and formulates a Hierarchical Distil-lation Learning (HDL) approach that aims to comprehensively distil the knowledge of a strong teacher model with strong learning capability to a lightweight student model with weak learning capability.
- Hallucinating multiple occluded face images of different resolutions24
A novel algorithm to super-resolve face images given multiple partially occluded inputs at different lower resolutions is proposed, which can probabilistically align multiple input images at different resolutions and recursively infer the high-resolution face image.
- Learning Support Vector Machines for A Multi-View Face Model24
This work extends SVMs to model the appearance of human faces which undergo nonlinear change across multiple views and uses inherent factors in the nature of the input images and the SVM classification algorithm to perform both multi-view face detection and pose estimation.
- Activity Based Video Content Trajectory Representation and Segmentation23
A novel approach is developed to segment continuous CCTV recordings according to the activities captured in the scene, which is more robust to noise and computationally much more efficient compared to existing on-line segmentation algorithms.
- View-based adaptive affine tracking23
This work proposes a model for view-based adaptive affine tracking of moving objects that avoids the need for feature-based matching in establishing correspondences through learning landmarks using an effective bootstrapping process based on colour segmentation and selective attention.
- Evaluating Feature Importance for Re-identification22
This study presents a comprehensive comparison and evaluation of up-to-date imagery features for person re-identification and introduces an unsupervised approach to learning a bottom-up measurement of feature importance.
- 22
- Minimum Cuts of A Time-Varying Background22
A novel background model is demonstrated exhibiting results superior to those achievable with the latter technique alone, and especially suitable for bac kground modelling in outdoor situations where variable lighting conditions p revail.
- 22
- A Unifying Theory of Active Discovery and Learning21
This study shows with systematic theoretical analysis how the previously disparate tasks of active discovery and learning can be cleanly unified into a single problem, and is able for the first time to develop a unified query algorithm to directly optimise this problem.
- Finding Rare Classes: Adapting Generative and Discriminative Models in Active Learning21
A unified active learning model to jointly discover new categories and learn to classify them; and a classifier combination algorithm that switches generative and discriminative classifiers as learning progresses that demonstrates the superiority of this approach over existing methods.
- Online Feature Selection Using Mutual Information for Real-Time Multi-view Object Tracking21
Since features in non-frontal views are computed on-the-fly by projective transforms under weak perspective projection, this framework requires only frontal-view training samples to track objects in multiple views.
- 21
- Deep learning prototype domains for person re-identification20
This work forms a deep learning based novel approach to automatic prototype-domain discovery for domain perceptive person re-id that outperforms state-of-the-art unsupervised methods significantly and is competitive against supervised methods which use labelled test domain data.
- Video Semantic Clustering with Sparse and Incomplete Tags20
This work develops a method for accurately clustering tagged videos based on a novel Hierarchical-MultiLabel Random Forest model capable of correlating structured visual and tag information and discovers more accurate semantic correlations among different video data, even with highly sparse/incomplete tags.
- 20
- Unsupervised Deep Learning via Affinity Diffusion19
A general-purpose unsupervised deep learning approach to deriving discriminative feature representations based on self-discovering semantically consistent groups of unlabelled training samples with the same class concepts through a progressive affinity diffusion process is introduced.
- 19
- 19
- 19
- Model Selection for Unsupervised Learning of Visual Context19
This study suggests that for learning visual context using a mixture model, BICr is the most appropriate criterion given sparse data, while CL-AIC should be chosen given moderate or large data sample sizes.
- 19
- Recognising Moving Faces19
An approach to engineering real-time computer vision systems for face recognition that involves recognition of multiple people in poorly constrained dynamic scenes and the need for integration of different processes using prediction and feedback is emphasised.
- Learning Rare Behaviours18
A novel topic model is proposed: Multi-Class Delta Latent Dirichlet Allocation which learns to model rare behaviours from a few weakly labelled videos as well as typical behaviours from uninteresting videos by collaboratively sharing features among all classes of footage.
- 17
- Data Driven Gesture Model Acquisition using Minimum Description Length17
It is shown that the generating processes for the atomic components and derived gesture models can be described by a mixture of Gaussian in their respective component and gesture space.
- Parallel computation of optic flow17
Both the tangential and normal components of the flow can be computed reliably where the image Hessian is well-conditioned and a fast algorithm to propagate flow along contours from such locations is proposed.
- Quantifying Temporal Saliency16
This work examines and proposes an extension to Kadir and Brady’s Scale Saliency Algorithm for quantifying temporal saliency and performing automatic spatial and temporal scale selection.
- 15
- Discovering visual concept structure with sparse and incomplete tags15
This work develops a method for modelling the inherent visual data concept structures based on a novel Hierarchical-Multi-Label Random Forest model capable of correlating structured visual and tag information so as to more accurately interpret the visual semantics.
- 15
- Learning Intrinsic Video Content Using Levenshtein Distance in Graph Partitioning15
The graph partitioning method is extended and in particular, the Normalised Cut model originally introduced for static image segmentation is extended to unsupervised clustering of temporal trajectories withfully automated model order selection.
- On the visual expectations of moving objects15
A Hidden Markov Model is used to represent the “hidden” regularities behind the apparent object movement in a scene and reproduce such regularities as “blind” expectations by adapting the weights on beliefs with new visual evidence.
- Group Association: Assisting Re-identification by Visual Context14
A novel people group representation and a group matching algorithm are proposed that addresses changes in the relative positions of people in a group and the latter uses the proposed group descriptors for measuring the similarity between two candidate images.
- 14
- Optimising dynamic graphical models for video content analysis14
It is demonstrated that CL-AIC is superior to the other scoring functions in building dynamic graphical models for solving two challenging problems in video content analysis: (1) content based surveillance video segmentation and (2) discovering causal/temporal relationships among visual events for group activity modelling.
- People detection in low-resolution video with non-stationary background14
A framework for robust people detection in low resolution image sequences of highly cluttered dynamic scenes with non-stationary background is presented and it is shown that the model is more robust with better detection rate compared to the model of Viola et al.
- From Local Temporal Correlation to Global Anomaly Detection14
The results demonstrate that the framework is able to detect not only anomalies that are visually obvious, but also those that are ambiguous or supported only by very weak visual evidence, e.g. those that can be easily missed by a human observer.
- 14
- Understanding visual behaviour14
One of the most noticeable developments in computer vision over the last decade has been the rapidly growing interest in the challenging problems of modelling and understanding human actions and behaviours captured in image sequences.
- Visual observation as reactive learning14
The use of a Hidden Markov Model to generate in time the probabilistic expectations of an observed object is described and some possible initial applications of such a framework for providing selective attention in visual observation are discussed.
- 13
- Exploring Semantic Attributes from A Foundation Model for Federated Learning of Disjoint Label Spaces13
A new Federated Zero-Shot Learning (FZSL) paradigm is formulated to learn mid-level semantic knowledge at multiple local clients with non-shared local data and cumulatively aggregate a globally generalised central model for deployment.
- Multi-perspective cross-class domain adaptation for open logo detection13
This work considers a more scalable open logo detection problem where only a fraction of logo classes are fully labelled whilst the remaining classes are only annotated with a clean icon image (e.g. 1-shot icon supervised).
- 13
- 12
- Multi-Modal Multi-Platform Person Re-Identification: Benchmark and Method12
The MP-ReID benchmark is introduced, a novel dataset designed specifically for multi-modality and multi-platform ReID, and UniPrompt ReID, a framework with specific-designed prompts, tailored for cross-modality and cross-platform scenarios.
- Learning Robust Graph Regularisation for Subspace Clustering12
This work identifies two critical limitations of the graph regularisation term employed in existing subspace clustering models and provides solutions for both of them and proposes to learn the graph and integrate the graph learning into the proposed l1-norm graph regularised optimisation problem.
- Real-time tracking for an integrated face recognition system12
A real-time tracking system suitable for surveillance applications was implemented using a symmetric spatio-temporal filter in order to determine regions of interest and normal components of visual motion were obtained at moving ‘edge’ features and Kalman filtering techniques were used for robust object tracking.
- 12
- Temporal Score Analysis for Understanding and Correcting Diffusion Artifacts11
This work proposes ASCED (Abnormal Score Correction for Enhancing Diffusion), that detects artifacts by monitoring abnormal score dynamics during the diffusion process, with a trajectory-aware on-the-fly mitigation strategy that appropriate generation of noise in the detected areas.
- 11
- 11
- Learning human pose in crowd11
A novel framework for locating people and inferring their body and head poses is proposed and significantly outperforms a state-of-the-art person detector and is able to yield extremely accurate head and body pose estimation in crowded public spaces.
- 11
- The dynamics of linear combinations: tracking 3D skeletons of human subjects11
A general framework for addressing three fundamental issues using linear combinations, the properties of the examples to linearly combine, the constraints, and the method for estimating the linear combinations coefficients for reconstructing an object based on noisy and incomplete visual observations is proposed.
- Exploiting Context in Gesture Recognition11
This work investigates how head pose can be exploited to impose contextual constraints for the recognition of gestures and demonstrates the method in a video conferencing scenario involving three people who use gestures to control the camera and focus of attention.
- Fusion of Perceptual Cues using Covariance Estimation11
A framework for fusing different information sources through estimation of covariance from observations is proposed and demonstrated in a face and 3D pose tracking system that fuses similarity-to-prototypes measures and skin colour to track head pose and face position.
- 10
- Decentralised Person Re-Identification with Selective Knowledge Aggregation10
A new Selective Knowledge Aggregation approach to decentralised person Re-ID to optimise the trade-off between model personalisation and generalisation and introduces a dual local normalisation mechanism to learn generalised normalisation layers in each local model, which are then transmitted to the global model for central aggregation.
- 10
- Translating topics to words for image annotation10
Compared to the original translation model, the visual topics learned by the probabilistic latent semantic analysis (PLSA) approach provide an intermediate abstract level of visual description.
- Online Video Behaviour Abnormality Detection Using Reliability Measure10
An approach is proposed for robust online behaviour recognition and abnormality detection based on discovering natural grouping of bebaviour patterns through unsupervised learning and a time accumulative reliability measure.
- 10
- AIM-Fair: Advancing Algorithmic Fairness via Selectively Fine-Tuning Biased Models with Contextual Synthetic Data9
This paper proposes AIM-Fair, a fine-tuning paradigm starting from a biased model initially trained on real-world data without demographic annotations that improves model fairness while maintaining utility, outperforming both fully and partially fine-tuned approaches to model fairness.
- Mitigate Domain Shift by Primary-Auxiliary Objectives Association for Generalizing Person ReID9
This work introduces a Primary-Auxiliary Objectives Association (PAOA) mechanism to calibrate the loss gradients of the auxiliary task towards the primary learning task gradients, and extends the model to form the PAOA+, which performs on-the-fly optimization against the auxiliary objective in order to maximize the model’s generative capacity in the test target domain.
- 9
- 9
- 9
- Corresponding dynamic appearances9
This paper extends the non-linear shape models to dynamic appearance models of both shape and texture across views and shows that Kernel PCA constrained the dynamic appearance model and eases model fitting to novel images.
- 9
- Generative Video Diffusion for Unseen Novel Semantic Video Moment Retrieval8
A hybrid selection mechanism is introduced that integrates three quantitative metrics to selectively incorporate synthetic video moments (novel video hypotheses) as enlarged additions to the original source training data, whilst minimising potential detrimental noise or unnecessary repetitions in the novel synthetic videos harmful to VMR learning.
- 8
- Characteristic Regularisation for Super-Resolving Face Images8
A method that joins the advantages of conventional SR and UDA models for characteristics consistifying and image super-resolving by introducing Characteristic Regularisation (CR) between them and makes the model training more effective and computationally tractable.
- Scalable Multi-camera Tracking in a Metropolis8
This work presents the design rationale, implementational considerations and quantitative evaluation of a retrospective forensic tool known as Multi-Camera Tracking, developed for re-identifying and back-tracking individuals within huge quantities of open-world CCTV video data sourced from a large distributed multi-camera network encompassing different public transport hubs in a metropolis.
- Surveillance video behaviour profiling and anomaly detection8
The results demonstrate that the framework is able to detect not only anomalies that are visually obvious, but also those that are ambiguous or supported only by very weak visual evidence, e.g. those that can be easily missed by a human observer.
- Capturing Correlations Among Facial Parts for Facial Expression Analysis8
A Matrix-based Canonical Correlation Analysis (MCCA) is proposed for better correlation analysis on 2D image or matrix data in general and compared to the traditional CCA, models more accurately correlations among image data with more compact representation using much fewer canonical factors.
- Coupling Face Registration and Super-Resolution8
A multi-resolution patch tensor based model to automatically super-resolve and register low-resolution testing face images and demonstrates the robustness of the approach on learning-based face super-resolution on real images.
- 8
- 8
- 8
- Feature-Distribution Perturbation and Calibration for Generalized Reid7
This work proposes a Feature-Distribution Perturbation and Calibration (PECA) method to derive generic feature representations for person ReID and designs a global calibration mechanism to align feature distributions across all the source domains to improve the model’s generalization capacity by eliminating domain bias.
- Person Re-Identification by Ranking Ensemble Representations7
This work addresses the problem of breach of deployment due to ignoring the probe-specific matching information against the gallery set encoded in ranking lists by exploring the idea of RANkinG Ensembles (RANGE) that learns such information from the ranking lists.
- Cross-domain traffic scene understanding by motion model transfer7
To provide cross-domain interpretation without risk of dramatic negative transfer, this paper introduces and forms a scene association criterion to quantify transferability of motion models from one scene to another.
- Behaviour in Context7
A range of established topics and emerging trends in visual analysis of behaviour from understanding facial expression, body gesture, action and intent, to the analysis of group activity, crowd and distributed behaviour, and gaining holistic awareness are introduced.
- Multi-layered Decomposition of Recurrent Scenes7
This work characterize a scene by estimating autocovariance of self-similarity, used previously in the context gait recognition, and introduces a spatio-temporal grid of histograms built in accordance with some chosen feature, which is then used to classify objects found in subsequent test data.
- Visual inference of human emotion and behaviour7
This work extends Canonical Correlation Analysis to multimodality correlation for bebaviour inference using both facial and body gestures and model temporal correlation among human movement patterns in a wider space using a mixture of Multi-Observation Hidden Markov Model for human behaviour profiling and behavioural anomaly detection.
- Segmenting Highly Textured Nonstationary Background7
By using the conditional probabilities yielded by pairwise cooccurrence of 4-connected pixels, and casting the problem as one of Combinatorial Optimization, the results show that useful segmentation is possible from challenging dynamic backgrounds.
- From Contextual Knowledge to Computational Constraints7
This work proposes a scheme that links scene-oriented contextual knowledge with the computational constraints required in visual motion segmentation and tracking, and discusses the experimental results from applying this method in improving existing techniques in traffic surveillance applications.
- 6
- Identity Alignment by Noisy Pixel Removal.6
This work forms a method of automatically detecting and removing identity class irrelevant pixels in auto-detected bounding boxes by learning to eliminate less informative pixels in identity alignment.
- A Unified Bayesian Framework for Adaptive Visual Tracking6
A unified generative model for multi-sensory adaptive tracking which cleanly integrates tracking and the modeling of appearance change across multiple features in the same framework is derived.
- 6
- 6
- Automated Detection of Localised Visual Events Over Varying Temporal Scales6
In this work, a general semantics-free method is proposed for the extraction of visual events based on the definition that an event is any sort of visual change, and results are shown on an artificial shopping scenario.
- 6
- Appearance-based face recognition under large head rotations in depth6
The computational difficulties in recognising faces undergoing large pose variations are described and the effectiveness of different linear appearance-based methods for the problem is evaluated.
- Guest Editorial Introduction to the Special Issue on Large-Scale Visual Sensor Networks: Architectures and Applications5
In a sensor network, a key role is played by largescale visual networks: the decreasing cost of smart devices enables sensor networks to scale to a large number of devices.
- Image and Video Understanding in Big Data5
It is clear that big visual data analysis and unerstanding impose significant scientific and technological chalenges since one not only should have to advance methods able to racefully scale to both big and diverse data whilst being computaionally cheap, but also should need to develop processes suitable or learning from big andiverse data without incurring prohibitive osts in human and monetary resources.
- 5
- Comparing Visual Feature Coding for Learning Disjoint Camera Dependencies5
The concept that visual features should be coded and selected automatically for robust and accurate time-delayed dependency learning is examined, and the Time Delayed Mutual Information (TDMI) is adopted due to its reported effectiveness and simplicity.
- Exploiting Periodicity in Recurrent Scenes5
A method of ensuring that the spatio-temporal model maintains synchronization with learned scene activity in spite of short-term breakdown in the reliability of acquired data, and long-term change of the mean fundamental period is described.
- 5
- 5
- Spotting Scene Change for Indexing Surveillance Video5
This work addresses the problem of spotting scene change by performing temporal segmentation on long video sequences with little colour information and observed content by employing sustained temporal change computed for local neighbourhoods in the image frames and defines a frame activity similarity metric.
- OpenUS: A Fully Open-Source Foundation Model for Ultrasound Image Analysis via Self-Adaptive Masked Contrastive Learning4
This paper proposes OpenUS, the first reproducible, open-source ultrasound foundation model built on a large collection of public data, and introduces a novel self-adaptive masking framework that combines contrastive learning with masked image modeling.
- Hybrid-Learning Video Moment Retrieval across Multi-Domain Labels4
A multiplE branch Video-text Alignment model (EVA) is introduced that performs cross-modal (visual-textual) matching information sharing and multi-modal feature alignment to optimise domain-invariant visual and textual features as well as per-task discriminative joint video-text representations.
- 4
- Zero-Shot Action Recognition by Word-Vector Embedding.4
This study constructs a mapping between visual features and a semantic descriptor of each action category, allowing new categories to be recognised in the absence of any visual training data, and investigates a series of synergistic improvements to the standard ZSL pipeline.
- Unsupervised Selective Transfer Learning for Object Recognition4
A novel unsupervised transfer learning framework that utilises unlabelled auxiliary data to quantify and select the most relevant transferrable knowledge for recognising a target object class from the background given very limited training target samples is proposed.
- Action recognition with cascaded feature selection and classification4
This work proposes a novel action classification approach based on cascaded feature selection and classification that significantly outperforms standard multi-class classifiers in addressing more challenging action recognition problems encountered in an unconstrained environment.
- 4
- GraphThinker: Reinforcing Temporally Grounded Video Reasoning with Event Graph Thinking3
GraphThinker is proposed, a reinforcement finetuning-based method that constructs structural event-level scene graphs and enhances visual grounding to jointly reduce hallucinations in video reasoning.
- Uncertainty-quantified Rollout Policy Adaptation for Unlabelled Cross-domain Temporal Grounding3
A Data-Efficient Unlabelled Cross-domain Temporal Grounding method, from which a model is first trained on a labelled source domain, then adapted to a target domain using only a small number of unlabelled videos from the target domain, which eliminates the need for target annotation and keeps both computational and storage overhead low enough to run in real time.
- 3
- 3
- 3
- Benchmarking Robustness of Text-Image Composed Retrieval3
This paper performs the first robustness study and establishes three new diversified benchmarks for systematic analysis of text-image composed retrieval against natural corruptions in both vision and text and further probe textural understanding.
- Local-Global Associative Frame Assemble in Video Re-ID3
This work concurrently optimises a local aligned quality module that distinguishes the quality of each frame based on local alignments, and a global correlated quality (GCQ) module that estimates global appearance correlations.
- Regularising Knowledge Transfer by Meta Functional Learning3
Experiments show that meta functional learning for regularisation knowledge transfer can benefit improving FSL classifiers and adopt an Iterative Update strategy on MFL (MFL-IU), which improves knowledge transfer regularisation from MFL by progressively learning the functional regularisation in knowledge transfer.
- 3
- 3
- Space-time audio-visual speech recognition with multiple multi-class probabilistic support vector machines.3
The evaluation shows that the proposed approach outperforms the alternatives both in Terms of recognition accuracy and in terms of robustness.
- 3
- Towards behaviour-recognition-based video surveillance3
A novel approach to modelling and recognising complex activities involving simultaneous movement of multiple objects based on visual event detection and reasoning instead of object tracking and trajectory matching is introduced.
- Wavelet-based holistic sequence descriptor for generating video summaries3
It is shown that an analysis of the trajectory shape is able to produce a temporal segmentation and an action based video summary is built using the Top Discriminative Active Pixels in the segments.
- Improved Local Flow3
The Curve Motion Constraint Equation provides an additional constraint to fully and locally estimate the flow field at seeds, which Brady calls seeds, in the parallel computation of optic flow.
- Large-Scale Pre-Trained Models Empowering Phrase Generalization in Temporal Sentence Localization2
A phrase-level Temporal Relationship Mining (TRM) framework that employs the temporal relationship between the phrase and the whole sentence to better understand each semantic entity in the sentence to improve phrase-level prediction quality and use phrase-level predictions to refine sentence-level ones.
- 2
- 2
- Rapid Person Re-Identification via Sub-space Consistency Regularization2
A novel Sub-space Consistency Regularization (SCR) algorithm that can speed up the ReID procedure by 0.25 times than real-value features under same dimensions whilst maintain a competitive accuracy, especially under short codes.
- 2
- 2
- Scaling up multi-camera tracking for real-world deployment2
A user-assisted multi-camera tracking system employing several key novel methodologies is developed further in order to improve tracking accuracy and further facilitate scalability to arbitrary numbers of camera views across much larger spatial areas and different locations.
- User-assisted visual search and tracking across distributed multi-camera networks2
A distributed multi-camera tracking system to visually aid human operators in associating people and objects effectively over multiple disjoint camera views in a large public space is described and an extensive evaluation of the greater effectiveness of the system as compared to existing approaches on industry-standard i-LIDS multi- camera data is provided.
- An Optimization Framework for Real-Time Appearance-Based Tracking under Weak Perspective2
It is shown that, under weak perspective projection, the projective warping of a rectangle feature can be approximated by a similarity transform with an additional free parameter.
- Outdoor Activity Recognition using Multi-Linked Temporal Processes2
The experiments demonstrate that the performance of a DML-HMM on modelling group activities in a noisy outdoor scene is superior compared to that of a Coupled Hidden Markov Model (CHMM).
- Face recognition from sequences using models of identity2
A method for modelling and recognising facial identity is described within the context of an integrated system for face recognition in dynamic scenes usingixture models.
- Visual behavior: modeling 'hidden' purposes in motion2
The concept of vision as behavior is related with results from some of the early studies on visual augmented hidden Markov model for representing `hidden' regularities in object motion and producing dynamic expectations of the moving object in the scene.
- 1
- ViSMaP: Unsupervised Hour-long Video Summarisation by Meta-Prompting1
The results show that ViSMaP achieves performance comparable to fully supervised state-of-the-art models while generalising across domains without sacrificing performance.
- 1
- 1
- A Framework of Meta Functional Learning for Regularising Knowledge Transfer1
This work proposes a novel framework of Meta Functional Learning by meta-learning a generalisable functional model from data-rich tasks whilst simultaneously regularising knowledge transfer to data-scarce tasks and shows that meta functional learning for knowledge transfer regularisation can improve FSL classifiers.
- 1
- 1
- Unsupervised Behaviour Profiling1
This chapter considers a clustering model that discovers the intrinsic grouping of behaviour patterns and detects anomalies on-line so that a decision on whether a behaviour pattern is normal is made as soon as sufficient visual evidence is collected without the completion of the observed pattern.
- Supervised Learning of Group Activity1
A contextual event based group activity representation, two different methods for activity based video segmentation, and a dynamic Bayesian network for supervised learning of an activity model with its model structure automatically learned from visual observations are described.
- Video Augmentation for Improving Audio Speech Recognition under Noise1
This work presents an audio-visual intermediate fusion strategy to locate speaker dependant pronounced digits in continuous video recorded with sound and shows that the proposed approach outperforms the others both in recognition accuracy and robustness in coping with variations in probe sequences.
- Optimal Dynamic Graphs for Video Content Analysis1
This study proposes a Completed Likelihood AIC (CL-AIC) scoring function that differs from existing ones by optimising explicitly both the explanation and prediction capabilities of a model simultaneously.
- Analysis and modelling of faces and gestures : Second International Workshop, AMFG 2005, Beijing, China, October 16, 2005 : proceedings1
A Dynamic Classification Method to Detect Faces and Identify Facial Expression and a Binary Decision Tree Implementation of a Boosted Strong Classifier for Face Recognition.
- 1
- Using Colour for Robust Tracking and Segmentation1
This work uses colour mixture models for real-time colour-based object localisation, tracking and segmentation in dynamic scenes, motivated by applications such as virtual studios and recognition of human faces, gestures and actions.
- –
- –
- –
- –
- –
- Neuro-Symbolic Spatial Reasoning in Segmentation–
This work proposes Relational Segmentor (RelateSeg) to impose explicit spatial relational constraints by first order logic (FOL) formulated in a neural network architecture, enabling end-to-end learning of spatial-relationally consistent segmentation in OVSS.
- –
- –
- Transfer Learning for Protein Structure Classification and Function Inference at Low Resolution–
This study demonstrates that it is possible to make accurate predictions of protein fold taxonomy from structures determined at low ($>$3 Angstroms) resolution, using a deep convolutional neural network trained on high-resolution structures, and explores the relationship between the information content of the input image and the predictive power of the model.
- Multi-Source Video Summarisation.–
This paper presents a novel unsupervised framework to learn jointly from both visual and independently-drawn non-visual data sources for discovering meaningful latent structure of surveillance video data, and investigates ways to cope with discrepant dimension and representation whist associating these heterogeneous data sources.
- –
- Object Localization by Generative Graph Configuration–
This work utilizes topological properties of a graphical object model, a part-based generative model with each part representing a patch of an object image, to create a reliable and flexible graph structure that encodes local topological information of anobject appearance.
- Tensor Representations for Object Classification and Detection–
A novel method is proposed for robust detection and categori zation of people from a wide-area distance by appearance and action and a complete framework for pedestrian detection based on the SST combined with Hausdorff distance that is able to manage object’s descript ion of variable lengths is presented.
- –
- –
- Real-time dictionary based super-resolution of surveillance video streams and targets–
This work considers one particular contemporary method based on sparse coding and shows how, by relaxing some model constraints, it can be sped up significantly compared to the reference implementation, and thus approach real-time performance with visually indistinct reduction in fidelity.
- Hierarchical Behaviour Discovery–
This chapter describes a Markov clustering topic model designed for unsupervised modelling and on-line processing of multi-object spatio-temporal behaviours in crowded public scenes that draws on machine learning theories on probabilistic topic models and dynamic Bayesian networks to achieve a robust hierarchical modelling of behaviours and their dynamics.
- –
- Towards Modelling Behaviour–
This chapter introduces a statistical learning approach, in particular probabilistic graphical models, to underpinning the mechanism for behavioural interpretation, and overviews different learning strategies for building behaviour models, ranging from supervised learning, unsupervised learning, semisupervisedLearning, weakly supervised learning and active learning.
- Fusion of Motion and Appearance for Robust People Detection in Cluttered Scenes–
This work introduces a spatial pyramid Gaussian Mixture approach to model variations of long-term human motion information, which is computed via an improved background modeling using spatial motion constrains and people appearance is modeled by histograms of oriented gradients.
- –
- Man in the Loop–
This chapter describes an active learning model for seeking human feedback on model selected queries that is internal to the model rather than decided by the human observer, to ensure that not only the most relevant human feedback is sought, but also the model is not subject to human bias without data support.
- Understanding Facial Expression–
This chapter considers the problems of how to construct a suitable representation and design an effective classification model for both static image based and dynamic sequence based automatic facial expression recognition.
- Multi-camera Behaviour Correlation–
This chapter describes a framework for modelling a joined-up representation of a synchronised global space, within which local activities from different observational viewpoints can be analysed and interpreted holistically, and develops a suitable mechanism capable of discovering and quantifying unknown correlations in temporal ordering and temporal delays among different camera views.
- –
- Modelling Rare and Subtle Behaviours–
To eliminate the prohibitive manual labelling cost, both in time and inconsistency, required by traditional supervised methods, this chapter describes a weakly supervised framework, in which a user needs not, or cannot, explicitly locate the target behaviours of interest in the training video data.
- –
- Holistic video detection–
The challenges and recent progress towards building computer vision systems for holistic video detection in a distributed network of multiple cameras based on object localisation, categorisation and tagging from different views in highly cluttered scenes are highlighted.
- Detection of human interaction from a distance using salient body behaviour modelling–
It is shown how pose or configuration based models of the human body can provide a rich framework for modelling human-human interactive body behaviour even when body parts are occluded and the framework is suitable for extracting salient features from thehuman body where each part is represented by a few pixels in each image frame.
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.