PriOr-Flow: Enhancing Primitive Panoramic Optical Flow with Orthogonal View
This work addresses a domain-specific challenge in panoramic vision for applications like robotics and surveillance, offering an incremental improvement by integrating orthogonal views into existing methods.
The paper tackles the problem of severe distortions in panoramic optical flow estimation, especially in polar regions, by proposing PriOr-Flow, a dual-branch framework that leverages orthogonal views to enhance performance, achieving state-of-the-art results on public datasets.
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.