Wenqiang Zhang

31

Papers

136

Total Citations

Papers (31)

OneTracker: Unifying Visual Object Tracking with Foundation Models and Efficient Tuning

De-confounded Data-free Knowledge Distillation for Handling Distribution Shifts

Boosting Adversarial Transferability with Spatial Adversarial Alignment

Scoring, Remember, and Reference: Catching Camouflaged Objects in Videos

Dynamic Semantic-Aware Correlation Modeling for UAV Tracking

NeurIPS 2025arXiv

OpenVIS: Open-vocabulary Video Instance Segmentation

OUS: Bridging Scene Context and Facial Features to Overcome the Rigid Cognitive Problem

A User-Friendly Framework for Generating Model-Preferred Prompts in Text-to-Image

Out of Thin Air: Exploring Data-Free Adversarial Robustness Distillation

Pixel-level Semantic Correspondence through Layout-aware Representation Learning and Multi-scale Matching Integration

Attribute Surrogates Learning and Spectral Tokens Pooling in Transformers for Few-Shot Learning

AziNorm: Exploiting the Radial Symmetry of Point Cloud for Azimuth-Normalized 3D Perception

Sparse Instance Activation for Real-Time Instance Segmentation

FERV39k: A Large-Scale Multi-Scene Dataset for Facial Expression Recognition in Videos

Towards Practical Certifiable Patch Defense With Vision Transformer

TopFormer: Token Pyramid Transformer for Mobile Semantic Segmentation

Correspondence Transformers With Asymmetric Feature Learning and Matching Flow Super-Resolution

MISC210K: A Large-Scale Dataset for Multi-Instance Semantic Correspondence

CiCo: Domain-Aware Sign Language Retrieval via Cross-Lingual Contrastive Learning

Dual Path Learning for Domain Adaptation of Semantic Segmentation

LVOS: A Benchmark for Long-term Video Object Segmentation

Hierarchical Visual Categories Modeling: A Joint Representation Learning and Density Estimation Framework for Out-of-Distribution Detection

Improving Generalization in Visual Reinforcement Learning via Conflict-aware Gradient Agreement Augmentation

Efficient Decision-based Black-box Patch Attacks on Video Recognition

Shape Matters: Deformable Patch Attack

D2SP: Dynamic Dual-Stage Purification Framework for Dual Noise Mitigation in Vision-based Affective Recognition.

Synthesizing Near-Boundary OOD Samples for Out-of-Distribution Detection

General Compression Framework for Efficient Transformer Object Tracking

MetaAnchor: Learning to Detect Objects with Customized Anchors

Reading Relevant Feature from Global Representation Memory for Visual Object Tracking

Content-based Unrestricted Adversarial Attack