NFlowJS-GF (with extra inlier set: Vistas and Wilddash2)

Reported on 2 benchmarks across 1 task · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology2 results

Anomaly DetectiononFishyscapes L&F
AP· uses extra data· 2021-12-23
69.43
best: 69.8 (cDNP+OE)
SOTA
Dense Out-of-Distribution Detection by Robust Learning on Synthetic Negative Data arXiv:2112.12833
Anomaly DetectiononFishyscapes L&F
FPR95· uses extra data· 2021-12-23
2
best: 47.43 (Dirichlet DeepLab)
Dense Out-of-Distribution Detection by Robust Learning on Synthetic Negative Data arXiv:2112.12833