Metraipolis › archive › dc_75b63bb129a2

Dyna3: VLM-Guided Training-Free 4D Reconstruction via Depth Foundation Models — arXiv:2610.01286

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.01286
License (as archived): CC0-1.0
Archived: 2026-10-06T19:13:13.258Z
SHA-256 of the archived text: 8f7a76cea75d362c25b05d554afb922ed450d9fee98276a819e3d5abc0b53059
Stable URL of this page: https://metraipolis.com/documento/dc_75b63bb129a2

How to cite

Dyna3: VLM-Guided Training-Free 4D Reconstruction via Depth Foundation Models — arXiv:2610.01286 — arXiv (en), https://arxiv.org/abs/2610.01286 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_75b63bb129a2 (Metraipolis archive dc_75b63bb129a2, sha256 8f7a76cea75d362c25b05d554afb922ed450d9fee98276a819e3d5abc0b53059)

BibTeX

@misc{dc75b63bb129a2,
  title = {Dyna3: VLM-Guided Training-Free 4D Reconstruction via Depth Foundation Models — arXiv:2610.01286},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.01286},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_75b63bb129a2; sha256 8f7a76cea75d362c25b05d554afb922ed450d9fee98276a819e3d5abc0b53059},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.01286. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Recent depth foundation models like Depth Anything 3 (DA3) achieve remarkable multi-view depth estimation but assume static 3D scenes, limiting their applicability to real-world dynamic environments. Existing training-free 4D methods like Easi3R and VGGT4D rely on correspondence-trained backbones whose attention encodes cross-frame matching, a property absent in depth-only models like DA3. We present Dyna3, a training-free framework that extends DA3 for 4D dynamic scene reconstruction without any fine-tuning. Our key insight is that DA3's cross-view features, though trained only for depth consistency, implicitly encode motion-discriminative signals when combined with best-match feature search across frames. Its static surfaces find consistent matches globally, while dynamic objects cannot. We further adopt vision-language models (VLM) to automatically generate scene-specific semantic prompts for SAM 3, enabling precise instance-level segmentation that distinguishes which objects move from what objects exist. For reconstruction, we decouple the scene into a cross-frame aligned static background and per-frame dynamic point clouds. Experiments on four datasets demonstrate that Dyna3 surpasses correspondence-trained methods with +5.5pp J-Mean over state-of-the-art VGGT4D on dynamic object segmentation, while achieving up to 13x faster pose estimation and 3x faster 4D reconstruction with 4 to 8x lower memory. Dyna3 could therefore enable much denser temporal sampling that prior methods cannot support.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-02 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.01286 — 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 2026-10-06T17:37:36.771Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).