Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks159 from public data
- The Unmanned Aerial Vehicle Benchmark: Object Detection and Tracking1,076
This work constructs a new UAV benchmark focusing on complex scenarios with new level challenges, and is the first time to explore issues in unconstrained scenes comprehensively in UAV based real scenes.
- Siamese Box Adaptive Network for Visual Tracking993
SiamBAN views the visual tracking problem as a parallel classification and regression problem, and thus directly classifies objects and regresses their bounding boxes in a unified FCN, making SiamB Ban more flexible and general.
- 252
- The Unmanned Aerial Vehicle Benchmark: Object Detection, Tracking and Baseline247
A novel approach based on the so-called Context-aware Multi-task Siamese Network (CMSN) model that explores new cues in UAV videos by judging the consistency degree between objects and contexts and that can be used for SOT and MOT is proposed.
- 205
- 188
- 174
- 152
- Exploiting Completeness and Uncertainty of Pseudo Labels for Weakly Supervised Video Anomaly Detection143
This work designs a multi-head classification module (each head serves as a classifier) with a diversity loss to maximize the distribution differences of predicted pseudo labels across heads and devise an iterative uncertainty pseudo label refinement strategy, which improves not only the initial pseudo labels but also the updated ones obtained by the desired classifier in the second stage.
- 139
- 139
- Toward Unified Token Learning for Vision-Language Tracking136
A simple, flexible and effective vision-language (VL) tracking pipeline, termed MMTrack, which casts VL tracking as a token generation task, and the design without other prior modules avoids multiple sub-tasks learning and hand-designed loss functions, significantly reducing the complexity of VL tracking modeling.
- Learning to Filter: Siamese Relation Network for Robust Tracking134
A novel Siamese relation network is proposed, which introduces two efficient modules, i.e. Relation Detector (RD) and Refinement Module (RM), which performs in a meta-learning way to obtain a learning ability to filter the distractors from the background while RM aims to effectively integrate the proposed RD into theSiamese framework to generate accurate tracking result.
- Hierarchical Modular Network for Video Captioning111
A hierarchical modular network to bridge video representations and linguistic semantics from three levels before generating captions is proposed and shows that the proposed method performs favorably against the state-of-the-art models on the two widely-used benchmarks.
- Multi-label classification by exploiting local positive and negative pairwise label correlation105
A simple and effective Bayesian model for multi-label classification by exploiting Local positive and negative Pairwise Label Correlations, named LPLC is proposed.
- Weakly-Supervised Crowd Counting Learns from Sorting Rather Than Locations98
A weakly-supervised counting network is proposed, which directly regresses the crowd numbers without the location supervision, and a soft-label sorting network along with the counting network, which sorts the given images by their crowd numbers.
- Spatiotemporal CNN for Video Object Segmentation98
A unified, end-to-end trainable spatiotemporal CNN model for VOS, which consists of two branches, i.e., the temporal coherence branch and the spatial segmentation branch, which is designed to capture the dynamic appearance and motion cues of video sequences to guide object segmentation.
- Real-time Visual Object Tracking with Natural Language Description86
This work argues that conditioning on the natural language (NL) description of a target provides information for longer-term invariance, and thus helps cope with typical tracking challenges, and proposes a novel deep tracking-by-detection formulation that can take advantage of NL descriptions.
- VisDrone-VDT2018: The Vision Meets Drone Video Detection and Tracking Challenge Results81
A large-scale video object detection and tracking dataset, which consists of 79 video clips with about 1.5 million annotated bounding boxes in 33, 366 frames, and the evaluation protocol of the VisDrone-VDT2018 challenge and the results of the algorithms on the benchmark dataset are presented.
- 71
- Multi-Attention Network for Compressed Video Referring Object Segmentation69
A multi-attention network which consists of dual-path dual-att attention module and a query-based cross-modal Transformer module designed to extract effective representation from compressed data in three modalities, i.e., I-frame, Motion Vector and Residual is proposed.
- Rethinking Sampling Strategies for Unsupervised Person Re-Identification67
It is suggested that deteriorated over-fitting is an important factor causing poor performance, and enhancing statistical stability can rectify this problem, and a simple yet effective approach is proposed, termed group sampling, which gathers samples from the same class into groups.
- 65
- 65
- Towards Universal Modal Tracking With Online Dense Temporal Token Learning63
This work proposes a universal video-level modality-awareness tracking model with online dense temporal token learning (called UM-ODTrack), designed to support various tracking tasks, utilizing the same model architecture and parameters, and introduces two simple yet effective online dense temporal token association mechanisms.
- 47
- 45
- 44
- PL-ranking43
An efficient low-rank stochastic subgradient descent method is designed to solve the proposed optimization problem and the correlations between features and labels so that the relevance between the different modalities can be enhanced after mapping them into the common subspace.
- 42
- Object Localization under Single Coarse Point Supervision40
A coarse point refinement (CPR) approach, which to the best knowledge is the first attempt to alleviate semantic variance from the perspective of algorithm, which ensures training high-performance object localizer under coarse point supervision.
- 39
- 38
- Multi-View Multi-Label Learning With View-Label-Specific Features38
This paper first learns a low dimensional label-specific data representation for each data view and construct a multi-label classification model based on it by exploiting label correlations and view consensus, and learns the contribution weight of each dataView to multi- label learning task for all the class labels jointly.
- 35
- Conditional GAN based individual and global motion fusion for multiple object tracking in UAV videos35
A Conditional Generative Adversarial Networks (GAN) based model is proposed to predict complex motions in UAV videos and robust motion prediction and improved MOT performance are achieved compared with state-of-the-art methods.
- 35
- Semantic-aware SAM for Point-Prompted Instance Segmentation34
A Semantic-Aware Instance Segmentation Network (SAPNet) that integrates Multiple Instance Learning with matching capability and SAM with point prompts is devised that strategically selects the most representative mask proposals generated by SAM to supervise segmentation, with a specific focus on object category information.
- 33
- 31
- Adaptively Unified Semi-supervised Learning for Cross-Modal Retrieval30
A novel dragging technique is fused into a unified linear regression model to make each sample moving to the direction of its relevant label while far away from that of its irrelevant ones, and an efficient optimization algorithm is developed to learn a discriminative common subspace for different modalities.
- MambaLCT: Boosting Tracking via Long-term Context State Space Model29
The advantage of MambaLCT is its ability to continuously extend the length of the context, capturing complete target change cues, which enhances the stability and robustness of the tracker, and enhances the model's ability to perceive targets in complex scenarios.
- A Tale of HodgeRank and Spectral Method: Target Attack Against Rank Aggregation Is the Fixed Point of Adversarial Game27
The effectiveness of the suggested target attack strategies is demonstrated by a series of toy simulations and several real-world data experiments, showing that the proposed methods could achieve the attacker's goal and the leading candidate of the perturbed ranking list is the designated one by the adversary.
- Cascade Cross-modal Attention Network for Video Actor and Action Segmentation from a Sentence27
The proposed algorithm allows the sentence to learn the most discriminative features of the video, remarkably improving the accuracy of matching and segmentation and significantly improves the performance over recent state-of-the-art methods.
- 26
- 25
- 25
- 24
- 24
- Progressive Multi-Resolution Loss for Crowd Counting23
This paper proposes to predict the density map at one resolution but measure its quality via a derived log-formed loss at multiple resolutions, and mathematically proves that the traditional single-resolution L2 loss is superior to a single-resolution L2 loss.
- Weakly Supervised Text-based Actor-Action Video Segmentation by Clip-level Multi-instance Learning23
This work is the first to perform AAVSS under weakly supervised situations and achieves comparable performance to state-of-the-art fully supervised approaches on two large-scale benchmarks, including A2D Sentences and J-HMDB Sentences.
- Multi-Networks Joint Learning for Large-Scale Cross-Modal Retrieval23
A novel deep framework of multi-networks joint learning for large-scale cross-modal retrieval that can simultaneously achieve specific features adapting to cross- modal task and learn a shared latent space for images and sentences is proposed.
- 23
- Spatial Self-Distillation for Object Detection with Inaccurate Bounding Boxes22
A Spatial Self-Distillation based Object Detector (SSD-Det) is heuristically proposed to mine spatial information to refine the inaccurate box in a self-distillation fashion and achieves state-of-the-art performance.
- Joint multi-view representation and image annotation via optimal predictive subspace learning22
This paper proposes to jointly conduct multi-view representation and image annotation via optimal predictive subspace learning, making the two tasks promote each other, and demonstrates that this method is superior to the other image annotation methods.
- 21
- Training Efficient Saliency Prediction Models with Knowledge Distillation21
This paper proposes a condensed and randomly initialized student network employed to achieve higher efficiency by transferring knowledge from complicated and well-trained teacher networks for efficient pixel-wise saliency prediction and proposes two statistical representations of feature maps as knowledge.
- 20
- Joint Multi-View Representation Learning and Image Tagging20
This paper presents an optimal predictive subspace learning method which jointly conducts multi-view representation learning and image tagging and introduces powerful predictors (SVM) for image tagging to achieve better annotation performance.
- Online web video topic detection and tracking with semi-supervised learning20
This paper forms topic detection and tracking as an online tracking, detection and learning problem, and obtains a topic tracker which could also discover novel topics from the new stream data.
- 20
- RETTA: Retrieval-enhanced test-time adaptation for zero-shot video captioning19
A novel zero-shot video captioning framework named Retrieval-Enhanced Test-Time Adaptation (RETTA), which takes advantage of existing pretrained large-scale vision and language models to directly generate captions with test-time adaptation.
- 19
- 19
- 18
- 18
- Release the Power of Online-Training for Robust Visual Tracking18
This paper proposes to improve the tracking accuracy via online training by squeezing redundant training data by analyzing the dataset distribution in low-level feature space and design statistic-based losses to increase the inter-class distance while decreasing the intra-class variance of high-level semantic features.
- Consistency-Aware Anchor Pyramid Network for Crowd Localization17
A Consistency-Aware Anchor Pyramid Network (CAAPN) comprising an Adaptive Anchor Generator (AAG) and a Localizer with Augmented Matching (LAM) that achieves favorable performance against state-of-the-art approaches on five challenging datasets.
- 16
- Beyond global fusion: A group-aware fusion approach for multi-view image clustering15
A group-aware multi-view fusion approach that is partitioned into groups according to their visual appearance, and different fusion weights are assigned to different groups, which achieves promising image clustering performance over the existing methods.
- Online selection of the best k-feature subset for object tracking15
This paper uses a special tree to formalize the feature subset space and uses conditional entropy to evaluate feature subset and a simple but efficient greedy search algorithm is developed to search this tree to obtain the optimal k-feature subset quickly.
- 14
- 14
- Fusing cross-media for topic detection by dense keyword groups14
This paper proposes a method of utilizing and fusing rich media information from web videos and news reports to extract weighted keyword groups, which are used for cross-media topic detection and demonstrates the effectiveness and efficiency of the proposed approach.
- Topic detection in cross-media: a semi-supervised co-clustering approach14
This work proposes a semi-supervised co-clustering approach for cross-media topic detection by a constrained non-negative matrix factorization and proves its mathematical rigorousness and correctness.
- 13
- Generalized Zero-Shot Video Classification via Generative Adversarial Networks12
This work proposes a description text dataset based on the UCF101 action recognition dataset, and proposes a new loss function that combines visual features with textual features that has a positive impact on the generalized zero-shot video classification.
- Localization-Aware Meta Tracker Guided With Adversarial Features12
This paper designs a novel intersection over union guided method to effectively balance the problem of classification and localization accuracy and creatively use adversarial features during offline training phase to improve the robustness of the classifier.
- 12
- Self-balance Motion and Appearance Model for Multi-object Tracking in UAV11
This paper proposes a novel self-balance method fusing appearance similarity and motion consistency and demonstrates the effectiveness of the proposed method with comparisons to several state-of-the-art trackers.
- 11
- 11
- Label Correlation Guided Deep Multi-View Image Annotation10
An image annotation method by integrating deep multi-view latent space learning and label correlation guided image annotation into a unified framework, which is termed as Label Correlation guided Deep Multi-view image annotation (LCDM) method is proposed.
- Cross-media topic detection associated with hot search queries10
This paper proposes an effective method for cross-media topic detection that considers using hot search queries as guidance to detect topics, and proposes an improved co-clustering method which can be well suited forCross-media data clustering.
- 9
- Weakly Supervised Video Individual Counting9
A weakly supervised VIC task is introduced, wherein trajectory labels are not provided, and two types of labels are provided to indicate traffic entering the field of view (inflow) and leaving the field view (outflow), and results show that the baseline weakly supervised method outperforms supervised methods.
- A Benchmark for UAV-View Natural Language-Guided Tracking9
A new benchmark, UAVNLT (Unmanned Aerial Vehicle Natural Language Tracking), for the UAV-view natural language-guided tracking task, achieving state-of-the-art performance and a baseline method suitable for this task is proposed, achieving state-of-the-art performance.
- Open-source deep learning-based automatic segmentation of mouse Schlemm's canal in optical coherence tomography images9
This work presents the first automatic method for segmentation of the SC lumen in OCT images of living mouse eyes, and the performance of the proposed model is comparable to the performance of a second human grader.
- 9
- CPR++: Object Localization via Single Coarse Point Supervision7
This study proposes coarse point refinement (CPR), which to the best knowledge is the first attempt to alleviate semantic variance from an algorithmic perspective in POL, and observes that CPR++ can obtain scale information and further reduce the semantic variance in a global region, thus guaranteeing high-performance object localization.
- 7
- 7
- Multi-Label Learning via Feature and Label Space Dimension Reduction7
This paper proposes a new framework for multi-label learning with a large number of class labels and features, i.e., Multi-Label Learning via Feature and Label Space Dimension Reduction, namely MLL-FLSDR, in which both the feature space and label space are reduced to low dimensional spaces respectively.
- Evidence-Grounded Trustworthy Multimodal Reasoning and Evaluation Benchmark in Complex Urban Scenes6
Evidence-grounded visual reasoning (EGVOR) is proposed, departing from implicit inference, EGVOR reformulates reasoning as the explicit generation of Evidence Atoms–structured triplets that enforce strict spatial-semantic alignment.
- 6
- ClickTrack: Towards real-time interactive single object tracking6
A new paradigm for single object tracking algorithms, ClickTrack, a new paradigm using clicking interaction for real-time scenarios, and the Guided Click Refiner (GCR), which accepts point and optional textual information as inputs, transforming the point into the bounding box expected by the operator.
- 6
- 6
- 6
- 5
- Dynamic Erasing Network Based on Multi-Scale Temporal Features for Weakly Supervised Video Anomaly Detection5
A Dynamic Erasing Network (DE-Net) is proposed for weakly supervised video anomaly detection, which learns multi-scale temporal features and designs a dynamic erasing strategy, which dynamically assesses the completeness of the detected anomalies and erases prominent abnormal segments in order to encourage the model to discover gentle abnormal segment in a video.
- 5
- 5
- Memory matrix5
A novel user experience for home video called Memory Matrix is presented, which could facilitate users to re-experience the joy of their memories, travelling along not only the time axis but also the space axis.
- 4
- 4
- Online learning affinity measure with CovBoost for multi-target tracking4
Experimental results on benchmark datasets demonstrate that tracklet affinity learned with the proposed CovBoosting method is more discriminative and could greatly improve the performance of the multi-target tracker.
- 4
- 4
- SDVPT: Semantic-Driven Visual Prompt Tuning for Open-world Object Counting3
This work proposes a plug-and-play Semantic-Driven Visual Prompt Tuning framework (SDVPT) that transfers knowledge from the training set to unseen categories with minimal overhead in parameters and inference time.
- 3
- 3
- 3
- Beyond appearance model: Learning appearance variations for object tracking3
A novel appearance variation prediction model which can be embedded into the existing generative appearance model based tracking framework and design three kinds of features which can encode motion, appearance, appearance reconstruction error information of target's surrounding image patches, and capture potential factors that may cause variations oftarget's appearance as well as its reconstruction error.
- P2Object: Single Point Supervised Object Detection and Instance Segmentation2
This paper introduces Point-to-Box Network (P2BNet), which constructs balanced instance-level proposal bags by generating proposals in an anchor-like way and refining the proposals in a coarse-to-fine paradigm, and largely surpasses the previous methods in terms of the mean average precision.
- 2
- 2
- 2
- 2
- Multi-Label Learning With Hidden Labels2
A novel approach named MLLHL is proposed to not only discover the hidden labels in the training data but also predict these hidden labels and observed labels for unseen examples simultaneously.
- 2
- 2
- Directly Locating Actions in Video with Single Frame Annotation1
This work proposes a novel method for point-supervised action localization that directly predicts actions' positions and length by a series of fully supervised action location iteratively.
- 1
- 1
- 1
- Tell Me What to Track.1
This work argues that conditioning on the natural language (NL) description of a target provides information for longer-term invariance, and thus helps cope with typical tracking challenges, and proposes a novel deep tracking-by-detection formulation that can take advantage of NL descriptions.
- 1
- 1
- 1
- 1
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- NRPerson: A Non-Registered Multi-Modal Benchmark for Tiny Person Detection and Localization–
A novel non-registered multi-modal benchmark named NRPerson is introduced, specifically designed to advance the field of tiny person detection and localization by accommodating the complexities of real-world scenarios.
- Towards Subject-Oriented Video Captioning via User-Specified Targets–
This work proposes a new video captioning task, subject-oriented video captioning, which allows users to specify the describing target via a bounding box, and constructs two subject-oriented video captioning datasets based on two widely used video captioning datasets: MSVD and MSRVTT.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- QoS-Based Services Selecting and Optimizing Algorithms on Grid–
This paper presents QoS-Based Services Selecting and Optimizing Composition on Grid, which designs service selecting and mapping algorithms for optimizing the model and creates an executed plan of optimum composition on Grid.
- –
- –
- –
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.