Nothing Special   »   [go: up one dir, main page]
Skip to content

Latest commit

 

History

History
43 lines (35 loc) · 1.72 KB

README.md

File metadata and controls

43 lines (35 loc) · 1.72 KB
Raw

MegaSaM: Accurate, Fast and Robust Structure and Motion from Casual Dynamic Videos

Project Page | Paper

arXiv 2024

Zhengqi Li1, Richard Tucker1, Forrester Cole1, Qianqian Wang1,2, Linyi Jin1,3, Vickie Ye2, Angjoo Kanazawa2, Aleksander Holynski1,2, Noah Snavely1,

1Google DeepMind, 2UC Berkeley, 3University of Michigan

TL;DR: MegaSaM estimates cameras and dense structure, quickly and accurately, from any static or dynamic video.

🚧 Please check back later.

Citation

 @article{li2024_MegaSaM,
  title     = {{MegaSaM}: Accurate, Fast and Robust Structure and Motion from Casual Dynamic Videos},
  author    = {Li, Zhengqi and Tucker, Richard and Cole, Forrester and Wang, Qianqian and Jin, Linyi and Ye, Vickie and Kanazawa, Angjoo and Holynski, Aleksander and Snavely, Noah},
  journal   = {arXiv preprint},
  year      = {2024}
}

Acknowledgment

Thanks to Rick Szeliski, Peter Hedman, Andrew Liu, Boyang Deng, and Lucy Chai for helpful proofreading, comments, and discussions.