SoccerNet-v2: A Dataset and Benchmarks for Holistic Understanding of Broadcast Soccer Videos

Benchmark Model Rank Results
action-spotting-on-soccernet-v2AudioVid (Vanderplaetse et al.)#9Average-mAP: 39.9
action-spotting-on-soccernet-v2NetVLAD (Giancola et al.)#10Average-mAP: 31.4
camera-shot-boundary-detection-on-soccernetHistogram (Scikit-Video)#1mAP: 78.5
camera-shot-boundary-detection-on-soccernetIntensity (Scikit-Video)#2mAP: 64.0
camera-shot-boundary-detection-on-soccernetContent (PySceneDetect)#3mAP: 62.2
camera-shot-boundary-detection-on-soccernetCALF (Cioppa et al.)#4mAP: 59.6