Výsledky vyhledávání

Report

AlignZeg: Mitigating Objective Misalignment for Zero-shot Semantic Segmentation

Autor: Ge, Jiannan, Xie, Lingxi, Xie, Hongtao, Li, Pandeng, Zhang, Xiaopeng, Zhang, Yongdong, Tian, Qi

Publikováno v: ECCV 2024

A serious issue that harms the performance of zero-shot visual recognition is named objective misalignment, i.e., the learning objective prioritizes improving the recognition accuracy of seen classes rather than unseen classes, while the latter is th

Externí odkaz: http://arxiv.org/abs/2404.05667

Zobrazit plný text záznamu

Report

Towards Balanced Alignment: Modal-Enhanced Semantic Modeling for Video Moment Retrieval

Autor: Liu, Zhihang, Li, Jun, Xie, Hongtao, Li, Pandeng, Ge, Jiannan, Liu, Sun-Ao, Jin, Guoqing

Video Moment Retrieval (VMR) aims to retrieve temporal segments in untrimmed videos corresponding to a given language query by constructing cross-modal alignment strategies. However, these existing strategies are often sub-optimal since they ignore t

Externí odkaz: http://arxiv.org/abs/2312.12155

Zobrazit plný text záznamu

Report

Dual-Stream Knowledge-Preserving Hashing for Unsupervised Video Retrieval

Autor: Li, Pandeng, Xie, Hongtao, Ge, Jiannan, Zhang, Lei, Min, Shaobo, Zhang, Yongdong

Unsupervised video hashing usually optimizes binary codes by learning to reconstruct input videos. Such reconstruction constraint spends much effort on frame-level temporal context changes without focusing on video-level global semantics that are mor

Externí odkaz: http://arxiv.org/abs/2310.08009

Zobrazit plný text záznamu

Report

Balanced Classification: A Unified Framework for Long-Tailed Object Detection

Autor: Qi, Tianhao, Xie, Hongtao, Li, Pandeng, Ge, Jiannan, Zhang, Yongdong

Conventional detectors suffer from performance degradation when dealing with long-tailed data due to a classification bias towards the majority head categories. In this paper, we contend that the learning bias originates from two factors: 1) the uneq

Externí odkaz: http://arxiv.org/abs/2308.02213

Zobrazit plný text záznamu

Report

MomentDiff: Generative Video Moment Retrieval from Random to Real

Autor: Li, Pandeng, Xie, Chen-Wei, Xie, Hongtao, Zhao, Liming, Zhang, Lei, Zheng, Yun, Zhao, Deli, Zhang, Yongdong

Video moment retrieval pursues an efficient and generalized solution to identify the specific temporal segments within an untrimmed video that correspond to a given language description. To achieve this goal, we provide a generative diffusion-based f

Externí odkaz: http://arxiv.org/abs/2307.02869

Zobrazit plný text záznamu