Metraipolis › archive › dc_bbdd91654dd7
Event-Driven Refresh and Recurrence Memory to Reduce Stale Grounding in Referring Video Object Segmentation — arXiv:2609.38758
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.38758
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T14:45:33.215Z
- SHA-256 of the archived text
87656484162d82ed4b5808981fe22f6d641d044e2bdb893cf9661f7c5f91259c- Stable URL of this page
https://metraipolis.com/documento/dc_bbdd91654dd7
How to cite
Event-Driven Refresh and Recurrence Memory to Reduce Stale Grounding in Referring Video Object Segmentation — arXiv:2609.38758 — arXiv (en), https://arxiv.org/abs/2609.38758 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bbdd91654dd7 (Metraipolis archive dc_bbdd91654dd7, sha256 87656484162d82ed4b5808981fe22f6d641d044e2bdb893cf9661f7c5f91259c)
BibTeX
@misc{dcbbdd91654dd7,
title = {Event-Driven Refresh and Recurrence Memory to Reduce Stale Grounding in Referring Video Object Segmentation — arXiv:2609.38758},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.38758},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bbdd91654dd7; sha256 87656484162d82ed4b5808981fe22f6d641d044e2bdb893cf9661f7c5f91259c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38758. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Referring Video Object Segmentation (RVOS) aims to produce a pixel-accurate mask sequence for an object specified by natural language. Sa2VA combines a multimodal large language model with SAM2 for grounded segmentation; however, its inference typically grounds the query from a small fixed set of initial keyframes and then relies on propagation. In long or dynamic videos, this can cause stale grounding and persistent false positives when the object composition changes (e.g., distractors enter or the target disappears/re-appears). We propose Event-Driven Refresh + Recurrence Memory (EDRRM), an enhancement that selectively re-invokes Sa2VA only at stable change points. EDRRM triggers refresh boundaries using an EMA-smoothed event score computed from tracking-derived cues (births/deaths and coarse composition/layout changes) with temporal constraints. A recurrence memory further retrieves anchor frames via CLIP similarity to re-condition the model on re-appearance events. Experiments on Ref-DAVIS17, MeViS, and ReVOS show that EDRRM achieves a competitive accuracy-efficiency trade-off relative to fixed-window and FrameDiff-SSIM baselines, maintaining comparable or superior J&F scores at substantially lower average refresh-call budgets and reducing false-positive failures. End-to-end runtime analysis further confirms that the overhead introduced by tracking, CLIP-based recurrence matching, and the identifiability gate remains modest relative to the dominant Sa2VA inference cost, thereby validating the efficiency of the proposed pipeline. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-01 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.38758 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T14:19:00.615Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).