Unsupervised Deep Graph Matching Based on Cycle Consistency
This addresses the problem of keypoint matching in computer vision without labeled data, offering a flexible and incremental improvement over existing methods.
The paper tackles unsupervised deep graph matching for keypoint matching in images without ground truth correspondences, achieving new state-of-the-art results.
We contribute to the sparsely populated area of unsupervised deep graph matching with application to keypoint matching in images. Contrary to the standard \emph{supervised} approach, our method does not require ground truth correspondences between keypoint pairs. Instead, it is self-supervised by enforcing consistency of matchings between images of the same object category. As the matching and the consistency loss are discrete, their derivatives cannot be straightforwardly used for learning. We address this issue in a principled way by building our method upon the recent results on black-box differentiation of combinatorial solvers. This makes our method exceptionally flexible, as it is compatible with arbitrary network architectures and combinatorial solvers. Our experimental evaluation suggests that our technique sets a new state-of-the-art for unsupervised graph matching.