Metraipolis › archive › dc_05cd3d362a57

TT-VidT: Decoupling the Temporal Axis for Efficient Motion-Centric Video Pretraining — arXiv:2609.33419

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.33419
License (as archived)
CC0-1.0
Archived
2026-10-07T23:51:36.271Z
SHA-256 of the archived text
1c642fc55c3ce42b6a9c385197e8bbdf7e998db7298f3a3b5d36a149617ad81a
Stable URL of this page
https://metraipolis.com/documento/dc_05cd3d362a57

How to cite

TT-VidT: Decoupling the Temporal Axis for Efficient Motion-Centric Video Pretraining — arXiv:2609.33419 — arXiv (en), https://arxiv.org/abs/2609.33419 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_05cd3d362a57 (Metraipolis archive dc_05cd3d362a57, sha256 1c642fc55c3ce42b6a9c385197e8bbdf7e998db7298f3a3b5d36a149617ad81a)
BibTeX
@misc{dc05cd3d362a57,
  title = {TT-VidT: Decoupling the Temporal Axis for Efficient Motion-Centric Video Pretraining — arXiv:2609.33419},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.33419},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_05cd3d362a57; sha256 1c642fc55c3ce42b6a9c385197e8bbdf7e998db7298f3a3b5d36a149617ad81a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33419. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Comparisons in video self-supervised learning often evaluate complete training recipes rather than isolating the method itself: architecture, objective, data exposure, schedule, scale, and decoder capacity can all vary at once. This makes it hard to identify which choices yield motion-prioritized representations, whose gains concentrate on frame-to-frame change while retaining useful appearance. We address this with a matched $4 \times 6 = 24$ architecture-objective study at roughly 170M ~ 190M encoder scale on $\sim$1.7M OpenVid and Moments-in-Time v2 clips for 8 epochs, and propose TT-VidT. TT-VidT combines a DINOv3-initialized ViT-B/16 per-frame spatial path with a compact Temporal Transfer Layer, trained by Diff Compression to reconstruct target frames from a first-frame appearance anchor and frame-specific motion tokens. The sweep shows that TT3D with Diff Compression, not either component alone, enters the strongest motion-sensitive regime, and decoder ablations favor a compact video-pretrained decoder. In final comparison, TT-VidT leads Jester, Something-Something V2, ARID, and Diving48 fine-tuning simultaneously, improving over the strongest non-TT row by 54% ~ 121%, while using 48% fewer encoder FLOPs than DisMo and 55% fewer than VideoMAE or V-JEPA2. HMDB51, IARD, and EPIC-Kitchens bound the claim.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.33419 — 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 2026-10-07T23:29:27.205Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).