Hybrid mamba for few-shot segmentation

• Main Authors: Xu, Qianxiong, Liu, Xuanyi, Zhu, Lanyun, Lin, Guosheng, Long, Cheng, Li, Ziyue, Zhao, Rui
• 其他作者: College of Computing and Data Science
• 格式: Conference or Workshop Item
• 語言: English
• 出版: 2025
• 主題: Computer and Information Science
• 在線閱讀: https://hdl.handle.net/10356/182607; http://arxiv.org/abs/2409.19613v1
• 機構: Nanyang Technological University

Many few-shot segmentation (FSS) methods use cross attention to fuse support foreground (FG) into query features, regardless of the quadratic complexity. A recent advance Mamba can also well capture intra-sequence dependencies, yet the complexity is only linear. Hence, we aim to devise a cross (a...