10 papers
DeCO: Discriminative Evidence Composition for Fine-Grained Dataset Distillation
Chuixuan Fan, Guang Li, Shijie Wang +5
Dataset distillation compresses a large training set into a compact synthetic set while preserving its downstream utility. However, existing methods primarily preserve global image…
FD: A Dedicated Framework for Fine-Grained Dataset Distillation
Hongxu Ma, Guang Li, Shijie Wang +5
Dataset distillation (DD) compresses a large training set into a small synthetic set, reducing storage and training cost, and has shown strong results on general benchmarks. Decoup…
Towards an Effective Action-Region Tracking Framework for Fine-grained Video Action Recognition
Baoli Sun, Yihan Wang, Xinzhu Ma +3
Fine-grained action recognition (FGAR) aims to identify subtle and distinctive differences among fine-grained action categories. However, current recognition methods often capture…
Referring Video Object Segmentation with Cross-Modality Proxy Queries
Baoli Sun, Xinzhu Ma, Ning Wang +2
Referring video object segmentation (RVOS) is an emerging cross-modality task that aims to generate pixel-level maps of the target objects referred by given textual expressions. Th…
Propagating Sparse Depth via Depth Foundation Model for Out-of-Distribution Depth Completion
Shenglun Chen, Xinzhu Ma, Hong Zhang +2
Depth completion is a pivotal challenge in computer vision, aiming at reconstructing the dense depth map from a sparse one, typically with a paired RGB image. Existing learning bas…
B-VLLM: A Vision Large Language Model with Balanced Spatio-Temporal Tokens
Zhuqiang Lu, Zhenfei Yin, Mengwei He +4
Recently, Vision Large Language Models (VLLMs) integrated with vision encoders have shown promising performance in vision understanding. The key of VLLMs is to encode visual conten…
Similarity Guided Multimodal Fusion Transformer for Semantic Location Prediction in Social Media
Zhizhen Zhang, Ning Wang, Haojie Li +1
Semantic location prediction aims to derive meaningful location insights from multimodal social media posts, offering a more contextual understanding of daily activities than using…
Is a Pure Transformer Effective for Separated and Online Multi-Object Tracking?
Chongwei Liu, Haojie Li, Zhihui Wang +1
Recent advances in Multi-Object Tracking (MOT) have demonstrated significant success in short-term association within the separated tracking-by-detection online paradigm. However,…
Learning Pixel-wise Continuous Depth Representation via Clustering for Depth Completion
Chen Shenglun, Zhang Hong, Ma XinZhu +2
Depth completion is a long-standing challenge in computer vision, where classification-based methods have made tremendous progress in recent years. However, most existing classific…
Towards Fair and Comprehensive Comparisons for Image-Based 3D Object Detection
Xinzhu Ma, Yongtao Wang, Yinmin Zhang +5
In this work, we build a modular-designed codebase, formulate strong training recipes, design an error diagnosis toolbox, and discuss current methods for image-based 3D object dete…