Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Stan Ziqing LiSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works154 from public data
- 2,146
- 1,521
- 931
- Learning Multi-scale Block Local Binary Patterns for Face Recognition682
Experiments on Face Recognition Grand Challenge (FRGC) ver2.0 database show that the proposed MB-LBP method significantly outperforms other LBP based face recognition algorithms.
- 659
- The Visual Object Tracking VOT2014 Challenge Results588
The Visual Object Tracking challenge 2014, VOT2014, aims at comparing short-term single-object visual trackers that do not apply pre-learned models of object appearance and introduces a new dataset with full annotation of targets by rotated bounding boxes and per-frame attribute.
- Salient Color Names for Person Re-identification550
This paper proposes a novel salient color names based color descriptor (SCNCD) to describe colors that outperforms the state-of-the-art performance (without user’s feedback optimization) on two challenging datasets (VIPeR and PRID 450S).
- Face Detection Based on Multi-Block LBP Representation447
This paper presents the use of a new set of distinctive rectangle features, called Multi-block Local Binary Patterns (MB-LBP), for face detection, which encodes rectangular regions' intensities by local binary pattern operator, and the resulting binary patterns can describe diverse local structures of images.
- 446
- 406
- 404
- 334
- Embedding Deep Metric for Person Re-identification: A Study Against Large Variations313
A novel moderate positive sample mining method to train robust CNN for person re-identification, dealing with the problem of large variation is proposed and the learning by a metric weight constraint is improved, so that the learned metric has a better generalization ability.
- Partial Face Recognition: Alignment-Free Approach313
An alignment-free face representation method based on Multi-Keypoint Descriptors (MKD), where the descriptor size of a face is determined by the actual content of the image, which is superior in recognizing both holistic and partial faces without requiring alignment.
- 307
- A Fast and Accurate Unconstrained Face Detector294
A new image feature called Normalized Pixel Difference (NPD) is proposed, computed as the difference to sum ratio between two pixel values, inspired by the Weber Fraction in experimental psychology, which is scale invariant, bounded, and able to reconstruct the original image.
- 283
- Learning Efficient Single-Stage Pedestrian Detectors by Asymptotic Localization Fitting256
An efficient single-stage pedestrian detection architecture (denoted as ALFNet) is designed, achieving state-of-the-art performance on CityPersons and Caltech, two of the largest pedestrian detection benchmarks, and hence resulting in an attractive pedestrian detector in both accuracy and speed.
- 222
- 207
- Unsupervised Domain Adaptation with Noise Resistible Mutual-Training for Person Re-identification186
A Noise Resistible Mutual-Training (NRMT) method, which maintains two networks during training to perform collaborative clustering and mutual instance selection, which outperforms the state-of-the-art UDA methods for person re-ID.
- 186
- Vehicle Re-Identification Using Quadruple Directional Deep Learning Features158
The proposed QD-DLF approach outperforms multiple state-of-the-art vehicle re-identification methods and is applicable to both VeRi and VehicleID databases.
- Heterogeneous Face Recognition from Local Structures of Normalized Appearance149
MB-LBP, an extension of LBP operator, is applied to encode the local image structures in the transformed domain, and further learn the most discriminant local features for recognition in heterogeneous face images.
- 147
- 138
- 136
- 136
- 135
- Anchor-Free Person Search132
This work presents the Feature-Aligned Person Search Network (AlignPS), the first anchor-free framework to efficiently tackle this challenging task, and proposes an aligned feature aggregation module to generate more discriminative and robust feature embeddings by following a "re-id first" principle.
- Graph Sampling Based Deep Metric Learning for Generalizable Person Re-Identification129
An efficient mini-batch sampling method, called graph sampling (GS), for large-scale deep metric learning, which improves the state of the art in generalizable person re-identification significantly and outperforms the competitive baseline.
- 119
- Soft-Margin Softmax for Deep Classification119
A novel soft-margin softmax (SM-Softmax) loss to improve the discriminative power of features that can not only adjust the desired continuous soft margin but also be easily optimized by the typical stochastic gradient descent (SGD).
- Pixel-Aware Deep Function-Mixture Network for Spectral Super-Resolution118
A pixel-aware deep function-mixture network for SSR, which is composed of a new class of modules, termed function- mixture (FM) blocks, which enables the network to pixel-wise determine the receptive field size and the mapping function.
- Surpassing Real-World Source Training Data: Random 3D Characters for Generalizable Person Re-Identification107
By training person re-identification models on these synthesized person images, it is demonstrated, for the first time, that models trained on virtual data can generalize well to unseen target images, surpassing the modelstrained on various real-world datasets.
- Multi-label convolutional neural network based pedestrian attribute classification107
The proposed multi-label convolutional neural network (MLCNN) can simultaneously predict multiple pedestrian attributes and significantly outperforms the SVM based method on the PETA database.
- 105
- 102
- Partial face recognition: An alignment free approach89
It is argued that a probe face image, holistic or partial, can be sparsely represented by a large dictionary of gallery descriptors by adopting a variable-size description which represents each face with a set of keypoint descriptors.
- Nighttime Face Recognition at Long Distance: Cross-Distance and Cross-Spectral Matching87
Experimental results show that the proposed face recognition algorithm outperforms two commercial state-of-the-art face recognition SDKs (FaceVACS and PittPatt) for long distance face recognition in both daytime and nighttime operations, highlighting the need for better data capture setup and robust face matching algorithms for cross spectral matching at distances greater than 100 meters.
- Deep Hybrid Similarity Learning for Person Re-Identification84
A deep hybrid similarity learning (DHSL) method based on a convolution neural network that is able to reasonably assign complexities of feature learning and metric learning in a CNN, so that the performance of person Re-ID is improved.
- Which photo groups should I choose? A comparative study of recommendation algorithms in Flickr73
This paper designs and compares seven Flickr group recommendation models and suggests that model-based approaches are beneficial compared with memory-based approaches in terms of top- k recommendation metric, and incorporating tags in the recommendation algorithms leads to an improvement of precision on the top 2% performance.
- Large Scale Similarity Learning Using Similar Pairs for Person Verification69
Results on the challenging datasets of face verification and person re-identification show that the proposed novel similarity measure outperforms the state-of-the-art methods and is able to handle heterogeneous data well.
- 68
- 66
- Center and Scale Prediction: Anchor-free Approach for Pedestrian and Face Detection65
This paper provides a new perspective where detecting objects is motivated as a high-level semantic feature detection task, and pedestrian and face detection is simplified as a straightforward center and scale prediction task through convolutions.
- Learning Anchored Unsigned Distance Functions with Gradient Direction Alignment for Single-view Garment Reconstruction65
AnchorUDF represents 3D shapes by predicting unsigned distance fields (UDFs) to enable open garment surface modeling at arbitrary resolution and achieves the state-of-the-art performance on single-view garment reconstruction.
- Attentive WaveBlock: Complementarity-Enhanced Mutual Networks for Unsupervised Domain Adaptation in Person Re-Identification and Beyond64
A novel light-weight module, the Attentive WaveBlock (AWB), which can be integrated into the dual networks of mutual learning to enhance the complementarity and further depress noise in the pseudo-labels is proposed.
- Attentive WaveBlock: Complementarity-enhanced Mutual Networks for Unsupervised Domain Adaptation in Person Re-identification.62
A novel light-weight module, the Attentive WaveBlock (AWB), which can be integrated into the dual networks of mutual learning to enhance the complementarity and further depress noise in the pseudo-labels is proposed.
- Exclusivity-Consistency Regularized Knowledge Distillation for Face Recognition61
This paper proposes a novel position-aware exclusivity to encourage large diversity among differents of the same layer to alleviate the low-capability of student network and investigates the prevailing knowledge for face recognition distillation to conclude that the knowledge of feature consistency is more flexible and preserves much more information than others.
- 56
- Realistic and Efficient Face Swapping: A Unified Approach with Diffusion Models52
This work proposes to reframe the face-swapping task as a self-supervised, train-time inpainting problem, enhancing the identity transfer while blending with the target image, and introduces a mask shuffling technique during inpainting training, which allows for a so-called universal model for swapping.
- AFAN: Augmented Feature Alignment Network for Cross-Domain Object Detection52
This work proposes a novel augmented feature alignment network (AFAN) which integrates intermediate domain image generation and domain-adversarial training into a unified framework and significantly outperforms the state-of-the-art methods on standard benchmarks for both similar and dissimilar domain adaptations.
- 51
- 49
- Flickr group recommendation based on tensor decomposition46
A tensor decomposition-based group recommendation model to suggest groups to users which can help tackle the problem of huge volume of groups in Flickr.
- Exploring Visual Context for Weakly Supervised Person Search45
This paper inventively considers weakly supervised person search with only bounding box annotations, and discovers three levels of context clues in unconstrained natural images that promote local and global discriminative capabilities and enhances clustering accuracy.
- 44
- Partial Face Matching between Near Infrared and Visual Images in MBGC Portal Challenge43
The proposed approach, without knowing statistical characteristics of the subjects or data, outperforms the methods of contrast significantly, with ten-fold higher verification rates at FAR of 0.1%.
- Deep Person Re-Identification with Improved Embedding.42
A new feature reweighting (FRW) layer is designed to explicitly emphasize the importance of each embedding dimension, thus leading to an improved embedding in a deep model for person re-identification.
- Deep person re-identification with improved embedding and efficient training42
A new feature reweighting (FRW) layer is designed to explicitly emphasize the importance of each embedding dimension, thus leading to an improved embedding in a deep model for person re-identification.
- 41
- Cloning Outfits from Real-World Images to 3D Characters for Generalizable Person Re-Identification39
An automatic approach is proposed to directly clone the whole outfits from real-world person images to virtual 3D characters, such that any virtual person thus created will appear very similar to its real- world counterpart.
- Face recognition using ordinal features39
An ordinal feature based method for face recognition using AdaBoost learning to select most effective hamming distance based weak classifiers and build a powerful classifier.
- 37
- Perceptual hash-based feature description for person re-identification37
The proposed method uses perceptual hashing to binarize low-level feature maps and combines several feature channels for feature encoding and three regional statistics are computed for hierarchical feature description.
- 37
- Towards accurate tiny vehicle detection in complex scenes35
A novel two-stage detector to perform tiny vehicle detection with high recall that consists of a backward feature enhancement network (BFEN) and a spatial layout preserving network (SLPN) is proposed.
- 35
- StoryAgent: Customized Storytelling Video Generation via Multi-Agent Collaboration34
StoryAgent is proposed, a versatile framework for video generation tasks, and novel techniques for preserving protagonist consistency are introduced, including a customized Image-to-Video (I2V) method, LoRA-BE, to enhance intra-shot temporal consistency, while a novel storyboard generation pipeline is proposed to maintain subject consistency across shots.
- Cross Dataset Person Re-identification34
A deep learning framework based on convolutional neural networks is presented to learn the person representation instead of existing hand-crafted features, and cosine metric is used to calculate the similarity.
- Cross-Modal Similarity Learning: A Low Rank Bilinear Formulation34
The proposed cross-modal similarity learning algorithm takes a bilinear formulation, and with the nuclear-norm penalization, it achieves low-rank representation and the accelerated proximal gradient algorithm is successfully imported to find the optimal solution with a fast convergence rate O(1/t2).
- ProtoDiff: Learning to Learn Prototypical Networks by Task-Guided Diffusion33
This work introduces ProtoDiff, a novel framework that leverages a task-guided diffusion model during the meta-training phase to gradually generate prototypes, thereby providing efficient class representations and the utilization of residual prototype learning, which leverages the sparsity of the residual prototype.
- DomainMix: Learning Generalizable Person Re-Identification Without Human Annotations33
A proposed domain-invariant feature learning method, called DomainMix, which can learn a person re-identification model from both synthetic and real-world data, for the first time, completely without human annotations, and achieves the current state of the art on several popular person re -identification datasets under direct cross-dataset evaluation.
- KD-DLGAN: Data Limited Image Generation via Knowledge Distillation32
KD-DLGAN is proposed, a knowledge-distillation based generation framework that introduces pre-trained vision-language models for training effective data-limited generation models and achieves superior image generation with limited training data.
- Efficient Person Search: An Anchor-Free Approach32
This work presents an anchor-free approach to efficiently tackling person search, and names its framework as Feature-Aligned Person Search Network (AlignPS), which achieves state-of-the-art or competitive performance on two challenging person search benchmarks, but can be also extended to other challenging searching tasks such as animal and object search.
- 32
- Face Recognition by Discriminant Analysis with Gabor Tensor Representation32
A 3rd-order Gabor tensor representation derived from a complete response set of Gabor filters across pixel locations and filter types is proposed, which shows promising results on FERET database.
- 30
- 30
- Any-Shift Prompting for Generalization Over Distributions29
This work proposes any-shift prompting: a general probabilistic inference framework that considers the relationship between training and test distributions during prompt learning, and introduces a transformer inference network with a pseudo-shift training mechanism.
- 29
- 28
- 28
- 28
- Pedestrian Detection: The Elephant In The Room27
This study finds that existing state-of-the-art pedestrian detectors generalize poorly from one dataset to another, and illustrates that diverse and dense datasets, collected by crawling the web, serve to be an efficient source of pre-training for pedestrian detection.
- Center and Scale Prediction: A Box-free Approach for Pedestrian and Face Detection27
This paper provides a new perspective where detecting objects is motivated as a high-level semantic feature detection task, and pedestrian and face detection is simplified as a straightforward center and scale prediction task through convolutions.
- Box Guided Convolution for Pedestrian Detection26
A box guided convolution (BGC) that can dynamically adjust the sizes of convolution kernels guided by the predicted bounding boxes is proposed, which provides position-aware receptive fields to address the challenge of large variations of scales.
- 26
- Pseudo-Labeling Based Practical Semi-Supervised Meta-Training for Few-Shot Learning25
A simple and effective meta-training framework, called pseudo-labeling based meta-learning (PLML), which effectively prevents the performance degradation of various FSL models under limited labeled data, and also significantly outperforms the representative SSMT models.
- Energy-Based Test Sample Adaptation for Domain Generalization24
A discriminative energy-based model is designed, which is trained on source domains to jointly model the conditional distribution for classification and data distribution for sample adaptation, and is optimized to simultaneously learn a classifier and an energy function.
- 24
- Human Parsing Based Texture Transfer from Single Image to 3D Human via Cross-View Consistency22
A human parsing based texture transfer model via cross-view consistency learning to generate the texture of 3D human body from a single image to reduce the appearance variation of human image and preserve the spatial distribution of semantic parts.
- 21
- 20
- 20
- Kernel sparse representation with pixel-level and region-level local feature kernels for face recognition19
This paper presents a kernel coordinate descent algorithm for the LASSO problem in the kernel space, and successfully integrates it in the SRC framework (called KCD-SRC) for face recognition, making it discriminative and robust against illumination variations and occlusions.
- Extracting non-negative basis images using pixel dispersion penalty19
A pixel dispersion penalty (PDP) is proposed, which effectively describes the spatial dispersion of pixels in an image without using any manually predefined structured patterns as constraints, and developed a spatial non-negative matrix factorization (Spatial NMF) and a spatialnon-negative component analysis (Sp spatial NCA).
- 18
- 18
- Joint Dimension Reduction and Metric Learning for Person Re-identification.18
This paper proposes a Retinex and maximum occurrence based feature representation method, which is robust to both illumination and viewpoint changes and comparable to the state of the art for person re-identification.
- Trends and Controversies17
The development of automata able to work effectively in conditions that are currently confined to visual surveillance, so called “recognition-in-the-wild,” is the development of automata able to work effectively in conditions that are currently confined to visual surveillance.
- 16
- Automatic Partial Face Alignment in NIR Video Sequences16
Experiments show the effectiveness of the proposed method for partial face alignment based on scale invariant feature transform, especially when PCA subspace, shape and temporal constraint are utilized.
- 16
- 15
- 15
- Feature Correlation Filter for Face Recognition15
This paper proposes a novel method, called "feature correlation filter (FCF)", by extending the concept of correlation filter to feature spaces, which preserves the benefits of conventional correlation filters, i.e., shift-invariant, occlusion-insensitive, and closed-form solution and also inherits virtues of the feature representations.
- POCE: Pose-Controllable Expression Editing13
POCE is presented, an innovative pose-controllable expression editing network that can generate realistic facial expressions and head poses simultaneously with just unpaired training images and can learn from unpaired face images effectively.
- Improve Pedestrian Attribute Classification by Weighted Interactions from Other Attributes11
This work proposes a novel pedestrian attribute classification method which exploits interactions among different attributes and is able to keep the balance of the independent decision score and interaction of other attributes to yield more robust classification results.
- TAL: Two-stream Adaptive Learning for Generalizable Person Re-identification9
This work argues that both domain-specific and domain-invariant features are crucial for improving the generalization ability of reid models, and designs a novel framework, which is name two-stream adaptive learning (TAL), to simultaneously model these two kinds of information.
- 9
- 9
- Interpretable and Generalizable Deep Image Matching with Adaptive Convolutions8
A model-free temporal cooccurrence based score weighting method is proposed, which improves the performance to a further extent, resulting in state-of-the-art results in cross-dataset evaluation.
- Learning Discriminative Features with Class Encoder8
The supervised information is incorporated to propose a novel formulation, namely class-encoder, whose training objective is to reconstruct a sample from another one of which the labels are identical, whose performance on benchmarks of classification and face recognition is improved.
- Face Classification: A Specialized Benchmark Study8
A specialized benchmark study in this paper focuses purely on face classification, and shows that, without the help of post-processing, the performance of face classification itself is still not very satisfactory, even with a powerful CNN method.
- 8
- Structured Ordinal Features for Appearance-Based Object Representation8
This paper extends SOF to Multi-scale Structured Ordinal Feature (MSOF) by concatenating binary strings of multi-scale SOFs at a fix position, which encodes not only microstructure but also macrostructure of image patterns, thus provides a more powerful image representation.
- A Discriminant Analysis Method for Face Recognition in Heteroscedastic Distributions6
The proposed LSR-LDA method improves the recognition accuracy over the conventional LDA by using the LSR step and the final projection matrix is obtained by multiply the pre-transformation matrix and the projective directions of LDA.
- 5
- 5
- Face Detection by Aggregating Visible Components4
A novel face detection method called Aggregating Visible Components (AVC) is proposed, which addresses pose variations and occlusions simultaneously in a single framework with low complexity.
- 4
- 3
- Self-Paced Learning for Open-Set Domain Adaptation3
This work proposes a novel framework based on self-paced learning to distinguish common and unknown class samples precisely, referred to as SPLOS (self- paced learning for open-set), and generates pseudo labels and design a cross-domain mixup method tailored for OSDA scenarios.
- 3
- 3
- Discriminant analysis with Gabor phase feature for robust face recognition3
An occlusion robust face image discriminant subspace is derived from Gabor phase difference representation, which adopts spectral regression–based discriminant analysis along with the extracted EGP features to find the most discriminating subspace for classification.
- 3
- 3
- Coarse-to-Fine Statistical Shape Model by Bayesian Inference3
This paper takes a predefined geometry shape as a constraint for accurate shape alignment and introduces Bayesian inference to make the whole shape more robust to local noise generated by the active shape, which leads to a compensation factor and a smooth factor for a coarse-to-fine shape search.
- Head Similarity: Modeling Structured Whole-Head Appearance Beyond Face Recognition2
This work develops a simple yet effective framework that jointly models identity discrimination and appearance-sensitive similarity through hierarchical supervision and identity-aware distillation and demonstrates the feasibility of structured whole-head similarity modeling.
- MP-ISMoE: Mixed-Precision Interactive Side Mixture-of-Experts for Efficient Transfer Learning2
This work proposes a novel Mixed-Precision Interactive Side Mixture-of-Experts framework (MP-ISMoE), which learns to select optimal experts by interacting with salient features from frozen backbones, thus suppressing knowledge forgetting and boosting performance.
- 2
- 2
- 2
- Learning Deep Semantic Embeddings for Cross-Modal Retrieval.2
A deep architecture called Deep Semantic Embedding (DSE) is proposed, which is trained in an end-to-end manner for image-text cross-modal retrieval, and the center loss is introduced beyond the commonly used softmax cross-entropy loss to achieve both inter-class separation and intraclass compactness.
- 2
- Memory-Efficient Transfer Learning with Fading Side Networks via Masked Dual Path Distillation1
This work develops a framework that enhances the performance by mutually distilling the frozen backbones and learnable side networks in fine-tuning, and discard the side network during inference without sacrificing accuracy, and designs a novel feature-based knowledge distillation method for the encoder structure with multiple layers.
- Large Models are Good Annotators for Zero-Shot Learning1
This work introduces a three-step interaction process with ChatGPT-comprising preliminary design, hierarchical refinement, and specific value determination-to generate attributes that are both category-shared and discriminative for classification, and develops a method that encodes attributes and their values as potential text pairings, leveraging CLIP's retrieval capabilities for annotation.
- DirectSwap: Paired, Mask-Free Video Head Swapping with Full-Reference Evaluation1
Di-rectSwap is proposed, a mask-free, direct video head-swapping framework that extends an image U-Net into a video diffusion model with a motion module and conditioning inputs and introduces the Motion-and Expression-Aware Reconstruction (MEAR) loss, which reweights the diffusion loss per pixel using frame-difference magnitudes and facial-landmark proximity, thereby enhancing cross-frame coherence in motion and expressions.
- 1
- 1
- Multimodal Biometrics Based on Near-Infrared Face Recognition1
This chapter contains sections titled: Introduction NIR Face-Based Multibiometrics Method of Multibiometricrics Fusion Experiments Conclusions Conclusions Acknowledgments References.
- FADRA: Frequency-Aware Diffusion with Residual Adaptation for Video Face Restoration–
FADRA, a frequency-aware diffusion framework with iterative residual adaptation specifically tailored for robust VFR, is proposed and a Frequency-Aware Loss is introduced that provides explicit supervision across multiple spectral bands, emphasizing visually sensitive frequency components that are crucial for perceptual quality and prone to temporal jittering.
- –
- –
- –
- –
- –
- –
- –
- Boosting Open-Set Domain Adaptation with Threshold Self-Tuning and Cross-Domain Mixup–
A novel threshold self- tuning and cross-domain mixup (TSCM) method that considers multiple criteria instead of only the confidence and uses the threshold generated by itself to separate common and unknown classes in the target domain.
- Similarity Scores Based Re-classification for Open-Set Person Re-identification–
To make the cross-view quadratic discriminant analysis metric learning method effectively learn both the projection matrix and the metric kernel with open-set data, an additional regularization factor is introduced to adjust the covariance matrix of the obtained subspace.
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.