Metraipolis › archive › dc_af693095fb18

Training-Free Temporal Abstraction for General Video Understanding — arXiv:2608.27929

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.27929
License (as archived)
CC0-1.0
Archived
2026-10-17T11:33:04.016Z
SHA-256 of the archived text
5865075bbda94eff73abcab22125b2afa0f78c16bf645bade3789fbb69496564
Stable URL of this page
https://metraipolis.com/documento/dc_af693095fb18

How to cite

Training-Free Temporal Abstraction for General Video Understanding — arXiv:2608.27929 — arXiv (en), https://arxiv.org/abs/2608.27929 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_af693095fb18 (Metraipolis archive dc_af693095fb18, sha256 5865075bbda94eff73abcab22125b2afa0f78c16bf645bade3789fbb69496564)
BibTeX
@misc{dcaf693095fb18,
  title = {Training-Free Temporal Abstraction for General Video Understanding — arXiv:2608.27929},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.27929},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_af693095fb18; sha256 5865075bbda94eff73abcab22125b2afa0f78c16bf645bade3789fbb69496564},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.27929. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Videos are expensive to analyze frame by frame, yet many video understanding tasks depend on knowing where relevant moments occur. A system may need to find when an action changes, locate the segment described by a sentence, or choose a few frames for a vision-language model. Existing methods often solve these problems separately, using task-specific training data or specialized architectures. We study whether a pretrained video-text model can provide enough temporal structure to support several of these tasks at once. We present STITCH, a training-free method that divides a video into semantically meaningful temporal chunks. STITCH embeds short video windows with a frozen video-text backbone and detects changes in the resulting embedding sequence. These chunks are computed once per video and reused across tasks. We evaluate STITCH on generic event boundary detection, language-based moment retrieval, and frame selection for long-video VLM reasoning. Across all three settings, STITCH remains competitive with more specialized methods while requiring no task-specific training, with especially clear gains when only a small number of frames or tokens can be processed. These results suggest that reusable temporal abstraction is a promising direction for general video understanding, allowing dense video streams to be converted once into semantic units that can be localized, retrieved, sampled, or reasoned over by downstream systems.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-31 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.27929 — 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 2026-10-17T11:10:05.282Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-31?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).