Re-Prompting SAM 3 via Object Retrieval: 3rd of the 5th PVUW MOSE Track
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917360891330560 |
|---|---|
| author | Gao, Mingqi Li, Sijie Han, Jungong |
| author_facet | Gao, Mingqi Li, Sijie Han, Jungong |
| contents | This technical report explores the MOSEv2 track of the PVUW 2026 Challenge, which targets complex semi-supervised video object segmentation. Built on SAM~3, we develop an automatic re-prompting framework to improve robustness under target disappearance and reappearance, severe transformation, and strong same-category distractors. Our method first applies the SAM~3 detector to later frames to identify same-category object candidates, and then performs DINOv3-based object-level matching with a transformation-aware target feature pool to retrieve reliable target anchors. These anchors are injected back into the SAM~3 tracker together with the first-frame mask, enabling multi-anchor propagation rather than relying solely on the initial prompt. This simple directly benefits several core challenges of MOSEv2. Our solution achieves a J&F of 51.17% on the test set, ranking 3rd in the MOSEv2 track. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_23788 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Re-Prompting SAM 3 via Object Retrieval: 3rd of the 5th PVUW MOSE Track Gao, Mingqi Li, Sijie Han, Jungong Computer Vision and Pattern Recognition This technical report explores the MOSEv2 track of the PVUW 2026 Challenge, which targets complex semi-supervised video object segmentation. Built on SAM~3, we develop an automatic re-prompting framework to improve robustness under target disappearance and reappearance, severe transformation, and strong same-category distractors. Our method first applies the SAM~3 detector to later frames to identify same-category object candidates, and then performs DINOv3-based object-level matching with a transformation-aware target feature pool to retrieve reliable target anchors. These anchors are injected back into the SAM~3 tracker together with the first-frame mask, enabling multi-anchor propagation rather than relying solely on the initial prompt. This simple directly benefits several core challenges of MOSEv2. Our solution achieves a J&F of 51.17% on the test set, ranking 3rd in the MOSEv2 track. |
| title | Re-Prompting SAM 3 via Object Retrieval: 3rd of the 5th PVUW MOSE Track |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.23788 |