Metraipolis › archive › dc_cee63e38cf61
UniFusion: Sparse-View 4D Reconstruction via Unified Spatio-temporal Depth Alignment — arXiv:2609.05888
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.05888
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T15:06:39.778Z
- SHA-256 of the archived text
e8334065d5b4f9489ac0aa468a8c4104b19fd016dcd0e07deb2f0cbf7ecd9caa- Stable URL of this page
https://metraipolis.com/documento/dc_cee63e38cf61
How to cite
UniFusion: Sparse-View 4D Reconstruction via Unified Spatio-temporal Depth Alignment — arXiv:2609.05888 — arXiv (en), https://arxiv.org/abs/2609.05888 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cee63e38cf61 (Metraipolis archive dc_cee63e38cf61, sha256 e8334065d5b4f9489ac0aa468a8c4104b19fd016dcd0e07deb2f0cbf7ecd9caa)
BibTeX
@misc{dccee63e38cf61,
title = {UniFusion: Sparse-View 4D Reconstruction via Unified Spatio-temporal Depth Alignment — arXiv:2609.05888},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.05888},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cee63e38cf61; sha256 e8334065d5b4f9489ac0aa468a8c4104b19fd016dcd0e07deb2f0cbf7ecd9caa},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.05888. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In this paper, we address the challenging problem of 4D reconstruction from sparse-view videos. This setup usually relies on monocular depth estimation to provide priors for the reconstruction model. A key challenge arises from limited cross-view overlap and temporal variation, making monocular depth predictions inconsistent across views and time. Existing methods align spatial and temporal dimensions in separate stages, requiring foreground segmentation masks while failing to leverage temporal cues for cross-view alignment. Contrary to these methods, we propose a unified spatial-temporal depth alignment framework that jointly resolves cross-view and cross-time inconsistencies without distinguishing foreground/background. Our method represents depth maps across views and time as a set of spatio-temporal neural fields. This representation not only yields fast convergence, but also captures spatio-temporal correlation among depth maps implicitly, without dependence on external segmentation/tracking models. We also propose a multi-view depth-order loss while leveraging the classic scale-and-shift-invariant loss to further improve the final depth quality. The aligned depths initialize and supervise Gaussian splatting models for 4D reconstruction. Experiments on Ego-Exo4D and EgoHuman demonstrate that our improved depth alignment substantially benefits dynamic Gaussian-splatting-based reconstruction methods for novel-time/view synthesis and geometry accuracy/consistency. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.05888 — 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 2026-10-16T14:49:57.010Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).