Rethinking Self-supervised Correspondence Learning: A Video Frame-level Similarity Perspective
This work addresses the challenge of generalizable correspondence learning for tasks like object tracking and segmentation, offering a novel approach that is incremental but shows strong gains.
The authors tackled the problem of learning representations for space-time correspondence in computer vision by proposing Video Frame-level Similarity (VFS) learning, which outperformed state-of-the-art self-supervised methods on OTB visual object tracking and DAVIS video object segmentation.
Learning a good representation for space-time correspondence is the key for various computer vision tasks, including tracking object bounding boxes and performing video object pixel segmentation. To learn generalizable representation for correspondence in large-scale, a variety of self-supervised pretext tasks are proposed to explicitly perform object-level or patch-level similarity learning. Instead of following the previous literature, we propose to learn correspondence using Video Frame-level Similarity (VFS) learning, i.e, simply learning from comparing video frames. Our work is inspired by the recent success in image-level contrastive learning and similarity learning for visual recognition. Our hypothesis is that if the representation is good for recognition, it requires the convolutional features to find correspondence between similar objects or parts. Our experiments show surprising results that VFS surpasses state-of-the-art self-supervised approaches for both OTB visual object tracking and DAVIS video object segmentation. We perform detailed analysis on what matters in VFS and reveals new properties on image and frame level similarity learning. Project page with code is available at https://jerryxu.net/VFS