Deep Graph Matching via Blackbox Differentiation of Combinatorial Solvers

25 Mar 2020  ·  Michal Rolínek, Paul Swoboda, Dominik Zietlow, Anselm Paulus, Vít Musil, Georg Martius ·

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

PDF Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Graph Matching PASCAL VOC BBGM-Multi F1 score 0.628 # 2
Graph Matching PASCAL VOC BBGM matching accuracy 0.801 # 9
F1 score 0.614 # 5
Graph Matching SPair-71k BBGM matching accuracy 0.8215 # 4
Graph Matching Willow Object Class BBGM matching accuracy 0.9718 # 9

Methods


No methods listed for this paper. Add relevant methods here