Metraipolis › archive › dc_6e4e4dd677e5

SM4RT: Cascaded Feed-forward Model for 4D Reconstruction — arXiv:2611.08436

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.08436
License (as archived)
CC0-1.0
Archived
2026-11-06T14:53:47.763Z
SHA-256 of the archived text
2dd6816b7e56e179be508a18d5b4418f95d08db129426198f8ea0cc6585b9220
Stable URL of this page
https://metraipolis.com/documento/dc_6e4e4dd677e5

How to cite

SM4RT: Cascaded Feed-forward Model for 4D Reconstruction — arXiv:2611.08436 — arXiv (en), https://arxiv.org/abs/2611.08436 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6e4e4dd677e5 (Metraipolis archive dc_6e4e4dd677e5, sha256 2dd6816b7e56e179be508a18d5b4418f95d08db129426198f8ea0cc6585b9220)
BibTeX
@misc{dc6e4e4dd677e5,
  title = {SM4RT: Cascaded Feed-forward Model for 4D Reconstruction — arXiv:2611.08436},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.08436},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6e4e4dd677e5; sha256 2dd6816b7e56e179be508a18d5b4418f95d08db129426198f8ea0cc6585b9220},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.08436. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We introduce SM4RT, a multi-task cascade feed-forward model for dynamic scene reconstruction from RGB videos. While existing feed-forward approaches typically address geometry reconstruction and motion estimation separately--and often predict motions in a single coordinate frame that must account for camera movement--SM4RT solves these tasks sequentially by first reconstructing 3D geometry and then bootstrapping motion estimation from the predicted structure. By expressing each pixel's 3D motion in its corresponding camera frame, SM4RT reduces ambiguity arising from camera motion. To address the sparsity of real-world motion labels, SM4RT further incorporates a weakly supervised loss that regularizes unlabeled point motions using labeled point positions at the other time step. We evaluate motion estimation on three dynamic-scene datasets--Kubric, Waymo, and Point Odyssey--and SM4RT achieves the best performance on most metrics compared with prior baselines. We additionally benchmark geometry reconstruction on five dynamic scene datasets and show that SM4RT achieves competitive performance with state-of-the-art methods. Overall, our unified framework demonstrates that combining geometry reconstruction and motion estimation in a cascade not only avoids performance trade-offs but also enhances motion estimation through improved geometric reasoning.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2611.08436 — 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 2026-11-06T14:25:42.630Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).