Abstract:Recent optical flow estimation methods often employ local cost sampling from a dense all-pairs correlation volume. This results in quadratic computational and memory complexity in the number of pixels. Although an alternative memory-efficient implementation with on-demand cost computation exists, this is significantly slower in practice and therefore many prior methods process images at downsampled resolutions, missing fine-grained details. To address this, we propose an algorithm for both memory and compute-efficient implementation of the all-pairs correlation volume sampling, still matching the exact mathematical operator as defined by RAFT. Our approach outperforms on-demand sampling by up to 92% while maintaining equally low memory usage, and performs at least on par with the default implementation with up to 99% lower memory usage. As cost sampling makes up a significant portion of the overall runtime, this can translate to up to 63% savings for the total end-to-end model inference on high-resolution inputs. Our evaluation of existing methods includes an 8K ultra-high-resolution dataset and an inference-time extension of the SEA-RAFT method. With this, we achieve state-of-the-art results at high resolutions both in accuracy and runtime.
| Comments: | CVPR 2026 |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG) |
| Cite as: | arXiv:2505.16942 [cs.CV] |
| (or arXiv:2505.16942v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2505.16942 arXiv-issued DOI via DataCite |
From: Karlis Martins Briedis [view email]
[v1]
Thu, 22 May 2025 17:30:38 UTC (3,245 KB)
[v2]
Tue, 26 May 2026 13:43:34 UTC (3,542 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。