Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Yihan, Lipson, Lahav, Deng, Jia
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2405.14793
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910458086162432
author Wang, Yihan
Lipson, Lahav
Deng, Jia
author_facet Wang, Yihan
Lipson, Lahav
Deng, Jia
contents We introduce SEA-RAFT, a more simple, efficient, and accurate RAFT for optical flow. Compared with RAFT, SEA-RAFT is trained with a new loss (mixture of Laplace). It directly regresses an initial flow for faster convergence in iterative refinements and introduces rigid-motion pre-training to improve generalization. SEA-RAFT achieves state-of-the-art accuracy on the Spring benchmark with a 3.69 endpoint-error (EPE) and a 0.36 1-pixel outlier rate (1px), representing 22.9% and 17.8% error reduction from best published results. In addition, SEA-RAFT obtains the best cross-dataset generalization on KITTI and Spring. With its high efficiency, SEA-RAFT operates at least 2.3x faster than existing methods while maintaining competitive performance. The code is publicly available at https://github.com/princeton-vl/SEA-RAFT.
format Preprint
id arxiv_https___arxiv_org_abs_2405_14793
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SEA-RAFT: Simple, Efficient, Accurate RAFT for Optical Flow
Wang, Yihan
Lipson, Lahav
Deng, Jia
Computer Vision and Pattern Recognition
We introduce SEA-RAFT, a more simple, efficient, and accurate RAFT for optical flow. Compared with RAFT, SEA-RAFT is trained with a new loss (mixture of Laplace). It directly regresses an initial flow for faster convergence in iterative refinements and introduces rigid-motion pre-training to improve generalization. SEA-RAFT achieves state-of-the-art accuracy on the Spring benchmark with a 3.69 endpoint-error (EPE) and a 0.36 1-pixel outlier rate (1px), representing 22.9% and 17.8% error reduction from best published results. In addition, SEA-RAFT obtains the best cross-dataset generalization on KITTI and Spring. With its high efficiency, SEA-RAFT operates at least 2.3x faster than existing methods while maintaining competitive performance. The code is publicly available at https://github.com/princeton-vl/SEA-RAFT.
title SEA-RAFT: Simple, Efficient, Accurate RAFT for Optical Flow
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.14793