Conventional video object segmentation (VOS) methods usually necessitate a substantial volume of pixel-level annotated video data for fully supervised learning. In this paper, we present HVC, a hybrid static-dynamic visual correspondence framework for self-supervised VOS. HVC extracts pseudo-dynamic signals from static...
Ya-Zhou Yao, Gen-Sheng Pei, Jian-Bo Jiao et al.· IEEE Transactions on Pattern...· 0 citations
Although Multimodal Large Language Models (MLLMs) have demonstrated impressive performance in Medical Visual Question Answering (Med-VQA), their reliance on global image features often lacks precise pixel-level grounding, thereby limiting clinical trustworthiness. To bridge the semantic gap between high-level clinical...
Hao-Wen Gu, Gen-Sheng Pei, Junxiong Mao et al.· 0 citations
This work proposes a task-specific pruning pipeline, named Cut-ViT, which first construct gram anchoring matrices from both spatial and semantic perspectives, and performs the subspace decomposition to extract the corresponding subspace bases.
Jianjian Yin, Liulei Li, Tao Chen et al.· 1 citation
A lightweight framework that leverages a memory bank to augment DCT-based low-frequency features and employs graph-enhanced cross-attention for effective visual-textual alignment is proposed, highlighting its efficiency and potential for clinical deployment.
Hao-Wen Gu, Gen-Sheng Pei, Ze-Ren Sun et al.· 2 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
