Learning Omnidirectional Flow in 360-degree Video via Siamese Representation
This work addresses the lack of benchmark datasets and adaptation challenges for omnidirectional video flow estimation, providing a new dataset and method for researchers in computer vision.
The paper tackles optical flow estimation in 360-degree videos by introducing FLOW360, the first perceptually natural-synthetic omnidirectional benchmark dataset with 40 videos and 4,000 frames, and proposes SLOF, a Siamese representation learning framework that achieves up to 40% performance improvement over state-of-the-art methods.
Optical flow estimation in omnidirectional videos faces two significant issues: the lack of benchmark datasets and the challenge of adapting perspective video-based methods to accommodate the omnidirectional nature. This paper proposes the first perceptually natural-synthetic omnidirectional benchmark dataset with a 360-degree field of view, FLOW360, with 40 different videos and 4,000 video frames. We conduct comprehensive characteristic analysis and comparisons between our dataset and existing optical flow datasets, which manifest perceptual realism, uniqueness, and diversity. To accommodate the omnidirectional nature, we present a novel Siamese representation Learning framework for Omnidirectional Flow (SLOF). We train our network in a contrastive manner with a hybrid loss function that combines contrastive loss and optical flow loss. Extensive experiments verify the proposed framework's effectiveness and show up to 40% performance improvement over the state-of-the-art approaches. Our FLOW360 dataset and code are available at https://siamlof.github.io/.