Deep Graph Matching via Blackbox Differentiation of Combinatorial Solvers
This work addresses graph matching for computer vision tasks, offering incremental improvements in performance and flexibility.
The paper tackles the problem of deep graph matching by integrating unmodified combinatorial solvers into an end-to-end trainable architecture, achieving state-of-the-art results on keypoint correspondence benchmarks.
Building on recent progress at the intersection of combinatorial optimization and deep learning, we propose an end-to-end trainable architecture for deep graph matching that contains unmodified combinatorial solvers. Using the presence of heavily optimized combinatorial solvers together with some improvements in architecture design, we advance state-of-the-art on deep graph matching benchmarks for keypoint correspondence. In addition, we highlight the conceptual advantages of incorporating solvers into deep learning architectures, such as the possibility of post-processing with a strong multi-graph matching solver or the indifference to changes in the training setting. Finally, we propose two new challenging experimental setups. The code is available at https://github.com/martius-lab/blackbox-deep-graph-matching