Metraipolis › archive › dc_98a26dee74fe

SAM2Dual: Training-Free, Dual Memory for Long-Term Video Object Segmentation — arXiv:2608.18640

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.18640
License (as archived)
CC0-1.0
Archived
2026-10-18T14:06:22.090Z
SHA-256 of the archived text
b58520237604b4ac5383666a1bb96bdb329e0622018cba0f645215e25d406817
Stable URL of this page
https://metraipolis.com/documento/dc_98a26dee74fe

How to cite

SAM2Dual: Training-Free, Dual Memory for Long-Term Video Object Segmentation — arXiv:2608.18640 — arXiv (en), https://arxiv.org/abs/2608.18640 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_98a26dee74fe (Metraipolis archive dc_98a26dee74fe, sha256 b58520237604b4ac5383666a1bb96bdb329e0622018cba0f645215e25d406817)
BibTeX
@misc{dc98a26dee74fe,
  title = {SAM2Dual: Training-Free, Dual Memory for Long-Term Video Object Segmentation — arXiv:2608.18640},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.18640},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_98a26dee74fe; sha256 b58520237604b4ac5383666a1bb96bdb329e0622018cba0f645215e25d406817},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.18640. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Long-term video object segmentation (VOS) remains challenging due to error accumulation under extended occlusions, re-appearance, and scene changes. Although SAM2 provides strong zero-shot performance, its streaming memory can amplify drift over long horizons when recent, unreliable predictions dominate the memory state. We propose SAM2Dual, a training-free, plug-and-play inference-time enhancement that improves long-video robustness without updating model weights. SAM2Dual introduces a Dual Memory design that explicitly separates (i) short-term memory for rapid local adaptation and (ii) long-term memory built via interval-based sampling to preserve global identity cues, combined through a gated fusion strategy. In addition, we present Text-Aware Memory (TAM), which extracts a compact word-level cue from early frames and uses text embeddings to reweight memory contributions based on semantic compatibility, supporting identity preservation when visual evidence becomes weak or ambiguous. Across long-term benchmarks, SAM2Dual consistently improves stability on long videos, raising J&F from 49.33 to 50.65 on MOSEv2 and achieving consistent gains on LVOSv2.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-20 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.18640 — 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 2026-10-18T14:00:22.567Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-20?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).