Few-Shot Referring Video Single- and Multi-Object Segmentation via Cross-Modal Affinity with Instance Sequence Matching
Year: 2025
Type: preprint
Abstract: Referring video object segmentation (RVOS) aims to segment objects in videos guided by natural language descriptions. We propose FS-RVOS, a Transformer-based model with two key components: a cross-modal affinity module and an instance sequence matching strategy, which extends FS-RVOS to multi-object segmentation (FS-RVMOS). Experiments show FS-RVOS and FS-RVMOS outperform state-of-the-art methods across diverse benchmarks, demonstrating superior robustness and accuracy.
Source: arXiv (Cornell University)
Authors: Liu, Heng, Li, Guanghui, Gao, Mingqi, Zhen, Xiantong, Zheng, Feng, Wang, Yang
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W6979347845
Cites:
Cited by:
Related to: