Pixel-wise Energy-biased Abstention Learning for Anomaly Segmentation on Complex Urban Driving Scenes

Benchmark Model Rank Results
anomaly-detection-on-fishyscapes-1PEBAL#3AP: 92.38FPR95: 1.73
anomaly-detection-on-fishyscapes-l-fPEBAL#6AP: 44.17FPR95: 7.58
anomaly-detection-on-lost-and-foundPEBAL#2AP: 78.29FPR: 0.81
anomaly-detection-on-road-anomalyPEBAL#6AP: 45.10FPR95: 44.58