Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks169 from public data
- 1,302
- Image-Image Domain Adaptation with Preserved Self-Similarity and Domain-Dissimilarity for Person Re-identification1,011
Through domain adaptation experiment, it is shown that images generated by SPGAN are more suitable for domain adaptation and yield consistent and competitive re-ID accuracy on two large-scale datasets.
- Conformer: Local Features Coupling Global Representations for Visual Recognition827
Experiments show that Conformer, under the comparable parameter complexity, outperforms the visual transformer (DeiT-B) by 2.3% on ImageNet and on MSCOCO, it outperforms ResNet-101 by 3.7% and 3.6% mAPs for object detection and instance segmentation, respectively, demonstrating the great potential to be a general backbone network.
- Prototype Mixture Models for Few-Shot Semantic Segmentation499
Proposed prototype mixture models (PMMs), which correlate diverse image regions with multiple prototypes to enforce the prototype-based semantic representation, incorporate rich channel-wised and spatial semantics from limited support images.
- 467
- SIXray: A Large-Scale Security Inspection X-Ray Benchmark for Prohibited Item Discovery in Overlapping Images363
A large-scale dataset and establish a baseline for prohibited item discovery in Security Inspection X-ray images, in which 6 classes of 8,929 prohibited items are manually annotated, and proposes an approach named class-balanced hierarchical refinement (CHR) to deal with these difficulties.
- Weakly Supervised Instance Segmentation Using Class Peak Response298
This paper addresses the challenging image-level supervised instance segmentation task by exploiting class peak responses to enable a classification network for instance mask extraction, and reports state-of-the-art results on popular benchmarks, including PASCAL VOC 2012 and MS COCO.
- Oriented Response Networks288
Over multiple state-of-the-art DCNN architectures, such as VGG, ResNet, and STN, it is consistently observed that replacing regular filters with the proposed ARFs leads to significant reduction in the number of network parameters and improvement in classification performance.
- 278
- C-MIL: Continuation Multiple Instance Learning for Weakly Supervised Object Detection261
A continuation optimization method is introduced into MIL and thereby creating continuation multiple instance learning (C-MIL), with the intention of alleviating the non-convexity problem in a systematic way.
- Beyond Bounding-Box: Convex-hull Feature Adaptation for Oriented and Densely Packed Object Detection258
- Min-Entropy Latent Model for Weakly Supervised Object Detection232
A min-entropy latent model (MELM) is proposed for weakly supervised object detection, unified with feature learning and optimized with a recurrent learning algorithm, which progressively transfers the weak supervision to object locations.
- Adaptive Keyframe Sampling for Long Video Understanding217
Experiments on two long video understanding benchmarks validate that AKS improves video QA accuracy (beyond strong baselines) upon selecting informative keyframes, and reveals the importance of information pre-filtering in video-based MLLMs.
- 211
- 205
- A configurable method for multi-style license plate recognition191
A new method for multi-style LP recognition by representing the styles with quantitative parameters, i.e., plate rotation angle, plate line number, character type and format, which will result in a faster and more precise recognition.
- 188
- 172
- Soft Proposal Networks for Weakly Supervised Object Localization152
This paper designs a network component, Soft Proposal (SP), to be plugged into any standard convolutional architecture to introduce the nearly cost-free object proposal, orders of magnitude faster than state-of-the-art methods.
- SRN: Side-Output Residual Network for Object Symmetry Detection in the Wild110
A new benchmark and an end-to-end deep learning approach are presented, opening up a promising direction for symmetry detection in the wild, and the proposed symmetry detection approach, named Side-output Residual Network (SRN), exploits the flow of errors among multiple scales to ease the problems of fitting complex outputs with limited layers.
- 105
- Visual abnormal behavior detection based on trajectory sparse reconstruction analysis100
A novel abnormal behavior detection approach by introducing trajectory sparse reconstruction analysis (SRA), solved by L1-norm minimization, leading to that a few of dictionary samples are used when reconstructing a behavior trajectory, which guarantees that the proposed approach is valid even when the dictionary set is very small.
- 95
- 84
- Visual object tracking via sample-based Adaptive Sparse Representation (AdaSR)83
A new approach for visual object tracking based on Sample-Based Adaptive Sparse Representation (AdaSR), which ensures that the tracked object is adaptively and compactly expressed with predefined samples, which is better than those of several representative tracking methods.
- 81
- Text detection and restoration in natural scene images76
A new method for text detection and recognition in natural scene images is presented, based on plane-to-plane homography, and results tested from a large dataset have demonstrated that the proposed method is effective and practical.
- 72
- 70
- 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.
- A novel laser vision sensor for weld line detection on wall-climbing robot63
Experiments show that the CSL sensor can capture 3D information of the weldment with very low measurement error, and the weld line detection approach is effective in wall-climbing robotic platform navigation.
- Anti-aliasing Semantic Reconstruction for Few-Shot Semantic Segmentation61
This paper reformulates few-shot segmentation as a semantic reconstruction problem, and converts base class features into a series of basis vectors which span a class-level semantic space for novel class reconstruction, which provides a systematic yet interpretable solution for few- shot learning problems.
- 55
- Vision-Dialog Navigation by Exploring Cross-Modal Memory54
The Cross-modal Memory Network (CMN) is proposed for remembering and understanding the rich information relevant to historical navigation actions and outperforms the previous state-of-the-art model on both seen and unseen environments.
- 53
- 50
- Agentic Reinforcement Learning with Implicit Step Rewards48
Implicit step rewards for agentic RL (iStar) are introduced, a general credit-assignment strategy that integrates seamlessly with standard RL algorithms without relying on additional rollouts or explicit step labels, and shows superior performance over frontier LLMs and strong RL baselines across domains.
- 47
- 45
- Pedestrian Detection with Deep Convolutional Neural Network44
This paper proposes to cascade simple Aggregated Channel Features (ACF) and rich Deep Convolutional Neural Network (DCNN) features for efficient and effective pedestrian detection in complex scenes.
- Fast-iTPN: Integrally Pre-Trained Transformer Pyramid Network With Token Migration43
Fast-iTPN can accelerate the inference procedure by up to up to 70% with negligible performance loss, demonstrating the potential to be a powerful backbone for downstream vision tasks.
- 43
- 42
- 40
- Combined feature evaluation for adaptive visual object tracking38
A combined feature evaluation approach in filter frameworks for adaptive object tracking by combining color histogram and gradient orientation histogram, which gives a representation of both color and contour.
- Generic-to-Specific Distillation of Masked Autoencoders37
This study proposes generic-to-specific distillation (G2SD), to tap the potential of small ViT models under the supervision of large models pretrained by masked autoencoders.
- Discretization-aware architecture search37
Discretization-aware architecture search is presented, with the core idea being adding a loss term to push the super-network towards the configuration of desired topology, so that the accuracy loss brought by discretization is largely alleviate.
- 37
- 34
- Vehicle Re-Identification with Location and Time Stamps31
This paper introduces an ensemble technique to combine the informative cues of multiple Re-ID models effectively and builds up a system to acquire the vehicle location and time stamps.
- ChatterBox: Multi-round Multimodal Referring and Grounding30
A new benchmark and an efficient vision-language model is presented, named ChatterBox, which outperforms existing models in MRG both quantitatively and qualitatively, paving a new path towards multimodal dialogue scenarios with complicated and precise interactions.
- 29
- 27
- Image captioning via semantic element embedding27
This paper integrates the CNN feature with the semantic feature into the proposed Element Embedding LSTM (EE-LSTM) model to predict a language description, which bridges the modality gap between visual images and semantic captions.
- 26
- 25
- Beyond Masking: Demystifying Token-Based Pre-Training for Vision Transformers24
This paper designs five other learning objectives that follow the same procedure as MIM but degrade the input image in different ways, and achieves superior performance over MIM in a series of downstream recognition tasks without extra computational cost.
- 24
- 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.
- 17
- 17
- 17
- Pedestrian detection in images via cascaded L1-norm minimization learning method17
The proposed CLML method, which is designed from the perspective of Vapnic's theory in the statistical learning, integrates feature selection with classifier construction via solving meaningful optimization models.
- 17
- Fault detection and classification in chemical processes using NMFSC and structural SVMs15
A novel fault detection and classification approach via Non‐negative Matrix Factorisation with Sparseness Constraints (NMFSC) and Structural Support Vector Machines (Structural SVMs) and the results show the superiority of the new method compared with the state‐of‐the‐art approaches.
- 14
- Depth-Guided Texture Diffusion for Image Semantic Segmentation13
A depth-guided texture diffusion approach that effectively tackles the outlined challenge of enhancing structural information vital for precisely extracting object outlines, and effectively bridges the disparity between depth and RGB modalities, enabling more accurate semantic segmentation.
- 13
- 13
- 12
- 12
- 11
- High performance person re-identification via a boosting ranking ensemble11
This paper proposes a novel framework for person re-identification called Boosting Ranking Ensemble (BRE), which adaptively assembles features and metrics from a ranking perspective and explores an effective rectifier loss in the proposed BRE framework to alleviate the influence of the misclassified samples.
- 11
- Unifying multimodal interactions for rumor diffusion prediction with global hypergraph modeling10
An HG2RLink framework of rumor diffusion prediction, which unifies multimodal interactions by global encoding and outperforms other methods with improvements ranging from 0.8% to 7.6% for the Hits@k metric and from 0.4% to 2.6% for the MAP@k metric across four diverse datasets.
- DQnet: Cross-Model Detail Querying for Camouflaged Object Detection10
It is argued that partial activation is caused by the intrinsic characteristics of CNN, where the convolution operations produce local receptive fields and experience difficulty to capture long-range feature dependency among image regions.
- 10
- 9
- Correspondence-Guided SfM-Free 3D Gaussian Splatting for NVS9
This study proposes a correspondence-guided SfM-free 3D Gaussian splatting for NVS that uses correspondences between the target and the rendered result to achieve better pixel alignment, facilitating the optimization of relative poses between frames.
- Spatial Preserved Graph Convolution Networks for Person Re-identification9
A graph convolution network based model for person re-identification is proposed to learn more discriminative feature embeddings, where a graph-structured relationship between person images and person parts are together integrated.
- Collective motion pattern inference via Locally Consistent Latent Dirichlet Allocation9
Experimental results on three public datasets indicate that LC-LDA based motion patterns can improve the trajectory clustering performance effectively and improve the unsupervised inference capability and compactness of learned collective motion patterns.
- Uncertainty-guided Optimal Transport in Depth Supervised Sparse-View 3D Gaussian8
This work introduces a novel method to supervise the depth distribution of 3D Gaussians, utilizing depth priors with integrated uncertainty estimates, and integrates a patch-wise optimal transport strategy to complement traditional L2 loss in depth supervision.
- 8
- 8
- 8
- 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
- SRN: Side-Output Residual Network for Object Reflection Symmetry Detection and Beyond7
Experimental results verify that the Sym-PASCAL benchmark is challenging related to real-world images, SRN achieves state-of-the-art performance, and MT-SRN has the capability to simultaneously predict edge and symmetry mask without loss of performance.
- A scalable convolutional neural network for task-specified scenarios via knowledge distillation7
A task-specified knowledge distillation algorithm is proposed to derive a simplified model with pre-set computation cost and minimized accuracy loss, which suits the resource constraint front-end systems well and proves the feasibility of the proposed approach as well as the existence of task- specified redundancy.
- 7
- 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.
- 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
- Rich Image Description Based on Regions6
A method of generating rich image descriptions from image regions by detecting regions with R-CNN and utilizing the RNN to generate sentences for image regions and proposing an optimization method to select one suitable region.
- 6
- Feature evaluation by particle filter for adaptive object tracking6
An online combined feature evaluation method for visual object tracking that adapts to changing appearances of both tracked object and background and extends the traditional filter framework from modeling motion states to modeling feature evaluation.
- 5
- 5
- 5
- 5
- 5
- 4
- 4
- 4
- 4
- 4
- 4
- A Belief Based Correlated Topic Model for Semantic Region Analysis in Far-Field Video Surveillance Systems4
Experiments show that the proposed belief based correlated topic model (BCTM) outperforms the state-of-the-art methods on qualitative results of learning semantic regions.
- Self-supervised feature-gate coupling for dynamic network pruning3
Experimental results validate that the proposed FGC method improves the baseline approach with significant margins, outperforming the state-of-the-arts with better accuracy-computation trade-off.
- 3
- 3
- 3
- 3
- 3
- Visual Object Tracking via One-Class SVM3
A new visual object tracking approach via one-class SVM (OC-SVM), inspired by the fact that OCSVM's support vectors can form a hyper-sphere, whose center can be regarded as a robust object estimation from samples, is proposed.
- 3
- Multi-posture Human Detection in Video Frames by Motion Contour Matching3
A thresholding method is used to classify a matching result into moving human or negatives by motion contour matching and in the matching process scale problem and interpersonal contour difference are considered.
- 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
- Visual trajectory analysis via Replicated Softmax-based models2
The Replicated Softmax model is applied to represent trajectories with “bag-of-word” features from a spatially distributed codebook and then used to characterize trajectory with latent topic units by stacking an additional label layer or representation layers.
- 2
- 2
- 2
- 2
- Fast and Robust Pedestrian Detection Algorithm with Multi-Scale Orientation Features2
The occluded features are completed with average features of training positives, given an occlusion model, which enable the proposed approach to work in crowd scenes.
- 2
- Image feature matching with network flow: a global optimization method2
A new approach is presented for obtaining feature matching based on the identified features from images by introducing the notion of a Min-cost K-flow Problem (MKP), which consists in finding a min-cost flow subject to the constraint that the flow value is K.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- Dimension Measuring System of Round Parts Based on Machine Vision1
Experimental results proved that the measuring scheme can realize real time detection with high accuracy and has broad prospects for the application of dimension measurement.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Exploring Complicated Search Spaces With Interleaving-Free Sampling–
This study explores the search algorithm upon a complicated search space with long-distance connections and shows that existing weight-sharing search algorithms fail due to the existence of interleaved connections, and presents a simple-yet-effective algorithm, termed interleaving-free neural architecture search (IF-NAS).
- 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.
- –
- –
- –
- Robust Visual Object Tracking via Sparse Representation and Reconstruction–
This paper proposes a new approach for robust visual object tracking via sparse representation and reconstruction, where two main contributions are devoted in terms of object representation and location respectively.
- –
- –
- Real-time pedestrian learning-tracking with information fusion–
The results of the experiments show good robustness and performance when facing extreme cases such as the object rotation, long-period occlusion, and high similarity between object and background.
- Visual object tracking via sparse reconstruction–
With the coefficient vector of the sparse representation, the tracked object is reconstructed in an instantaneous sample set, which improves the tracking adaptation to background variation, object shape change, and partial occlusion.
- –
- Shaking video stabilization with content completion–
A new stabilization algorithm to counterbalance the shaking motion in a video based on classical Kandade-Lucas- Tomasi (KLT) method is presented in this paper and experiments demonstrate the effectiveness of the proposed algorithm.
- Text detection in natural scene images with feature combination–
Experimental results on a large dataset show that the combination of features in different detection stages is competent for text detection task.
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 2026-10-16. 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.