Memory-Efficient Optical Flow via Radius-Distribution Orthogonal Cost Volume

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xu, Gangwei, Chen, Shujun, Jia, Hao, Feng, Miaojie, Yang, Xin
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929739789238272
author Xu, Gangwei
Chen, Shujun
Jia, Hao
Feng, Miaojie
Yang, Xin
author_facet Xu, Gangwei
Chen, Shujun
Jia, Hao
Feng, Miaojie
Yang, Xin
contents The full 4D cost volume in Recurrent All-Pairs Field Transforms (RAFT) or global matching by Transformer achieves impressive performance for optical flow estimation. However, their memory consumption increases quadratically with input resolution, rendering them impractical for high-resolution images. In this paper, we present MeFlow, a novel memory-efficient method for high-resolution optical flow estimation. The key of MeFlow is a recurrent local orthogonal cost volume representation, which decomposes the 2D search space dynamically into two 1D orthogonal spaces, enabling our method to scale effectively to very high-resolution inputs. To preserve essential information in the orthogonal space, we utilize self attention to propagate feature information from the 2D space to the orthogonal space. We further propose a radius-distribution multi-scale lookup strategy to model the correspondences of large displacements at a negligible cost. We verify the efficiency and effectiveness of our method on the challenging Sintel and KITTI benchmarks, and real-world 4K ($2160\!\times\!3840$) images. Our method achieves competitive performance on both Sintel and KITTI benchmarks, while maintaining the highest memory efficiency on high-resolution inputs.
format Preprint
id arxiv_https___arxiv_org_abs_2312_03790
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Memory-Efficient Optical Flow via Radius-Distribution Orthogonal Cost Volume
Xu, Gangwei
Chen, Shujun
Jia, Hao
Feng, Miaojie
Yang, Xin
Computer Vision and Pattern Recognition
The full 4D cost volume in Recurrent All-Pairs Field Transforms (RAFT) or global matching by Transformer achieves impressive performance for optical flow estimation. However, their memory consumption increases quadratically with input resolution, rendering them impractical for high-resolution images. In this paper, we present MeFlow, a novel memory-efficient method for high-resolution optical flow estimation. The key of MeFlow is a recurrent local orthogonal cost volume representation, which decomposes the 2D search space dynamically into two 1D orthogonal spaces, enabling our method to scale effectively to very high-resolution inputs. To preserve essential information in the orthogonal space, we utilize self attention to propagate feature information from the 2D space to the orthogonal space. We further propose a radius-distribution multi-scale lookup strategy to model the correspondences of large displacements at a negligible cost. We verify the efficiency and effectiveness of our method on the challenging Sintel and KITTI benchmarks, and real-world 4K ($2160\!\times\!3840$) images. Our method achieves competitive performance on both Sintel and KITTI benchmarks, while maintaining the highest memory efficiency on high-resolution inputs.
title Memory-Efficient Optical Flow via Radius-Distribution Orthogonal Cost Volume
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2312.03790