The 1st Solution for 7th LSVOS RVOS Track: SaSaSa2VA
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_ | 1866911219219169280 |
|---|---|
| author | Niu, Quanzhu Gong, Dengxian Chen, Shihao Zhang, Tao Zhou, Yikang Yuan, Haobo Qi, Lu Li, Xiangtai Ji, Shunping |
| author_facet | Niu, Quanzhu Gong, Dengxian Chen, Shihao Zhang, Tao Zhou, Yikang Yuan, Haobo Qi, Lu Li, Xiangtai Ji, Shunping |
| contents | Referring video object segmentation (RVOS) requires segmenting and tracking objects in videos conditioned on natural-language expressions, demanding fine-grained understanding of both appearance and motion. Building on Sa2VA, which couples a Multi-modal Large Language Model (MLLM) with the video segmentation model SAM2, we identify two key bottlenecks that limit segmentation performance: sparse frame sampling and reliance on a single [SEG] token for an entire video. We propose Segmentation Augmented and Selective Averaged Sa2VA (SaSaSa2VA) to address these issues. On the 7th LSVOS Challenge (RVOS track), SaSaSa2VA achieves a $\mathcal{J\&F}$ of 67.45, ranking first and surpassing the runner-up by 2.80 points. This result and ablation studies demonstrate that efficient segmentation augmentation and test-time ensembling substantially enhance grounded MLLMs for RVOS. The code is released in Sa2VA repository: https://github.com/bytedance/Sa2VA. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_16972 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | The 1st Solution for 7th LSVOS RVOS Track: SaSaSa2VA Niu, Quanzhu Gong, Dengxian Chen, Shihao Zhang, Tao Zhou, Yikang Yuan, Haobo Qi, Lu Li, Xiangtai Ji, Shunping Computer Vision and Pattern Recognition Artificial Intelligence Referring video object segmentation (RVOS) requires segmenting and tracking objects in videos conditioned on natural-language expressions, demanding fine-grained understanding of both appearance and motion. Building on Sa2VA, which couples a Multi-modal Large Language Model (MLLM) with the video segmentation model SAM2, we identify two key bottlenecks that limit segmentation performance: sparse frame sampling and reliance on a single [SEG] token for an entire video. We propose Segmentation Augmented and Selective Averaged Sa2VA (SaSaSa2VA) to address these issues. On the 7th LSVOS Challenge (RVOS track), SaSaSa2VA achieves a $\mathcal{J\&F}$ of 67.45, ranking first and surpassing the runner-up by 2.80 points. This result and ablation studies demonstrate that efficient segmentation augmentation and test-time ensembling substantially enhance grounded MLLMs for RVOS. The code is released in Sa2VA repository: https://github.com/bytedance/Sa2VA. |
| title | The 1st Solution for 7th LSVOS RVOS Track: SaSaSa2VA |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2509.16972 |