Enhancing Sa2VA for Referent Video Object Segmentation: 2nd Solution for 7th LSVOS RVOS Track
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914046989565952 |
|---|---|
| author | Hong, Ran Lu, Feng Cao, Leilei Yan, An Jiang, Youhai Zhu, Fengjie |
| author_facet | Hong, Ran Lu, Feng Cao, Leilei Yan, An Jiang, Youhai Zhu, Fengjie |
| contents | Referential Video Object Segmentation (RVOS) aims to segment all objects in a video that match a given natural language description, bridging the gap between vision and language understanding. Recent work, such as Sa2VA, combines Large Language Models (LLMs) with SAM~2, leveraging the strong video reasoning capability of LLMs to guide video segmentation. In this work, we present a training-free framework that substantially improves Sa2VA's performance on the RVOS task. Our method introduces two key components: (1) a Video-Language Checker that explicitly verifies whether the subject and action described in the query actually appear in the video, thereby reducing false positives; and (2) a Key-Frame Sampler that adaptively selects informative frames to better capture both early object appearances and long-range temporal context. Without any additional training, our approach achieves a J&F score of 64.14% on the MeViS test set, ranking 2nd place in the RVOS track of the 7th LSVOS Challenge at ICCV 2025. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_15546 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Enhancing Sa2VA for Referent Video Object Segmentation: 2nd Solution for 7th LSVOS RVOS Track Hong, Ran Lu, Feng Cao, Leilei Yan, An Jiang, Youhai Zhu, Fengjie Computer Vision and Pattern Recognition Referential Video Object Segmentation (RVOS) aims to segment all objects in a video that match a given natural language description, bridging the gap between vision and language understanding. Recent work, such as Sa2VA, combines Large Language Models (LLMs) with SAM~2, leveraging the strong video reasoning capability of LLMs to guide video segmentation. In this work, we present a training-free framework that substantially improves Sa2VA's performance on the RVOS task. Our method introduces two key components: (1) a Video-Language Checker that explicitly verifies whether the subject and action described in the query actually appear in the video, thereby reducing false positives; and (2) a Key-Frame Sampler that adaptively selects informative frames to better capture both early object appearances and long-range temporal context. Without any additional training, our approach achieves a J&F score of 64.14% on the MeViS test set, ranking 2nd place in the RVOS track of the 7th LSVOS Challenge at ICCV 2025. |
| title | Enhancing Sa2VA for Referent Video Object Segmentation: 2nd Solution for 7th LSVOS RVOS Track |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2509.15546 |