Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2506.23897 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908431134228480 |
|---|---|
| author | Liu, Longliang Feng, Miaojie Cheng, Junda Xiang, Jijun Zhu, Xuan Yang, Xin |
| author_facet | Liu, Longliang Feng, Miaojie Cheng, Junda Xiang, Jijun Zhu, Xuan Yang, Xin |
| contents | Panoramic optical flow enables a comprehensive understanding of temporal dynamics across wide fields of view. However, severe distortions caused by sphere-to-plane projections, such as the equirectangular projection (ERP), significantly degrade the performance of conventional perspective-based optical flow methods, especially in polar regions. To address this challenge, we propose PriOr-Flow, a novel dual-branch framework that leverages the low-distortion nature of the orthogonal view to enhance optical flow estimation in these regions. Specifically, we introduce the Dual-Cost Collaborative Lookup (DCCL) operator, which jointly retrieves correlation information from both the primitive and orthogonal cost volumes, effectively mitigating distortion noise during cost volume construction. Furthermore, our Ortho-Driven Distortion Compensation (ODDC) module iteratively refines motion features from both branches, further suppressing polar distortions. Extensive experiments demonstrate that PriOr-Flow is compatible with various perspective-based iterative optical flow methods and consistently achieves state-of-the-art performance on publicly available panoramic optical flow datasets, setting a new benchmark for wide-field motion estimation. The code is publicly available at: https://github.com/longliangLiu/PriOr-Flow. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_23897 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PriOr-Flow: Enhancing Primitive Panoramic Optical Flow with Orthogonal View Liu, Longliang Feng, Miaojie Cheng, Junda Xiang, Jijun Zhu, Xuan Yang, Xin Computer Vision and Pattern Recognition Panoramic optical flow enables a comprehensive understanding of temporal dynamics across wide fields of view. However, severe distortions caused by sphere-to-plane projections, such as the equirectangular projection (ERP), significantly degrade the performance of conventional perspective-based optical flow methods, especially in polar regions. To address this challenge, we propose PriOr-Flow, a novel dual-branch framework that leverages the low-distortion nature of the orthogonal view to enhance optical flow estimation in these regions. Specifically, we introduce the Dual-Cost Collaborative Lookup (DCCL) operator, which jointly retrieves correlation information from both the primitive and orthogonal cost volumes, effectively mitigating distortion noise during cost volume construction. Furthermore, our Ortho-Driven Distortion Compensation (ODDC) module iteratively refines motion features from both branches, further suppressing polar distortions. Extensive experiments demonstrate that PriOr-Flow is compatible with various perspective-based iterative optical flow methods and consistently achieves state-of-the-art performance on publicly available panoramic optical flow datasets, setting a new benchmark for wide-field motion estimation. The code is publicly available at: https://github.com/longliangLiu/PriOr-Flow. |
| title | PriOr-Flow: Enhancing Primitive Panoramic Optical Flow with Orthogonal View |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2506.23897 |