Deep Learning of Partial Graph Matching via Differentiable Top-K

Benchmark Model Rank Results
graph-matching-on-imcpt-sparsegm-100GCAN-AFAT-U#1F1 score: 0.715
graph-matching-on-imcpt-sparsegm-100GCAN-AFAT-I#2F1 score: 0.709
graph-matching-on-imcpt-sparsegm-100NGMv2-AFAT-U#3F1 score: 0.703
graph-matching-on-imcpt-sparsegm-100NGMv2-AFAT-I#4F1 score: 0.701
graph-matching-on-imcpt-sparsegm-100NGMv2#5F1 score: 0.676
graph-matching-on-imcpt-sparsegm-100PCA-GM#6F1 score: 0.575
graph-matching-on-imcpt-sparsegm-50GCAN-AFAT-I#1F1 score: 0.729
graph-matching-on-imcpt-sparsegm-50NGMv2-AFAT-I#2F1 score: 0.728
graph-matching-on-imcpt-sparsegm-50NGMv2-AFAT-U#3F1 score: 0.720
graph-matching-on-imcpt-sparsegm-50GCAN-AFAT-U#4F1 score: 0.711
graph-matching-on-imcpt-sparsegm-50NGMv2#5F1 score: 0.703
graph-matching-on-imcpt-sparsegm-50PCA-GM#6F1 score: 0.631
graph-matching-on-pascal-vocGCAN-AFAT-U#2F1 score: 0.620
graph-matching-on-pascal-vocGCAN-AFAT-I#3F1 score: 0.616
graph-matching-on-pascal-vocNGMv2-AFAT-U#5F1 score: 0.602
graph-matching-on-pascal-vocNGMv2-AFAT-I#7F1 score: 0.599
graph-matching-on-willow-object-classGCAN-AFAT-I#16F1 score: 0.837
graph-matching-on-willow-object-classNGMv2-AFAT-I#17F1 score: 0.831
graph-matching-on-willow-object-classGCAN-AFAT-U#18F1 score: 0.823
graph-matching-on-willow-object-classNGMv2-AFAT-U#19F1 score: 0.817