Shape of Motion: 4D Reconstruction from a Single Video

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Qianqian, Ye, Vickie, Gao, Hang, Zeng, Weijia, Austin, Jake, Li, Zhengqi, Kanazawa, Angjoo
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912652555452416
author Wang, Qianqian
Ye, Vickie
Gao, Hang
Zeng, Weijia
Austin, Jake
Li, Zhengqi
Kanazawa, Angjoo
author_facet Wang, Qianqian
Ye, Vickie
Gao, Hang
Zeng, Weijia
Austin, Jake
Li, Zhengqi
Kanazawa, Angjoo
contents Monocular dynamic reconstruction is a challenging and long-standing vision problem due to the highly ill-posed nature of the task. Existing approaches depend on templates, are effective only in quasi-static scenes, or fail to model 3D motion explicitly. We introduce a method for reconstructing generic dynamic scenes, featuring explicit, persistent 3D motion trajectories in the world coordinate frame, from casually captured monocular videos. We tackle the problem with two key insights: First, we exploit the low-dimensional structure of 3D motion by representing scene motion with a compact set of SE(3) motion bases. Each point's motion is expressed as a linear combination of these bases, facilitating soft decomposition of the scene into multiple rigidly-moving groups. Second, we take advantage of off-the-shelf data-driven priors such as monocular depth maps and long-range 2D tracks, and devise a method to effectively consolidate these noisy supervisory signals, resulting in a globally consistent representation of the dynamic scene. Experiments show that our method achieves state-of-the-art performance for both long-range 3D/2D motion estimation and novel view synthesis on dynamic scenes. Project Page: https://shape-of-motion.github.io/
format Preprint
id arxiv_https___arxiv_org_abs_2407_13764
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Shape of Motion: 4D Reconstruction from a Single Video
Wang, Qianqian
Ye, Vickie
Gao, Hang
Zeng, Weijia
Austin, Jake
Li, Zhengqi
Kanazawa, Angjoo
Computer Vision and Pattern Recognition
Monocular dynamic reconstruction is a challenging and long-standing vision problem due to the highly ill-posed nature of the task. Existing approaches depend on templates, are effective only in quasi-static scenes, or fail to model 3D motion explicitly. We introduce a method for reconstructing generic dynamic scenes, featuring explicit, persistent 3D motion trajectories in the world coordinate frame, from casually captured monocular videos. We tackle the problem with two key insights: First, we exploit the low-dimensional structure of 3D motion by representing scene motion with a compact set of SE(3) motion bases. Each point's motion is expressed as a linear combination of these bases, facilitating soft decomposition of the scene into multiple rigidly-moving groups. Second, we take advantage of off-the-shelf data-driven priors such as monocular depth maps and long-range 2D tracks, and devise a method to effectively consolidate these noisy supervisory signals, resulting in a globally consistent representation of the dynamic scene. Experiments show that our method achieves state-of-the-art performance for both long-range 3D/2D motion estimation and novel view synthesis on dynamic scenes. Project Page: https://shape-of-motion.github.io/
title Shape of Motion: 4D Reconstruction from a Single Video
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.13764