TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/PEBAL

PEBAL

Reported on 8 benchmarks across 1 task · 1 paper · 4 SOTA

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

Methodology8 results

  • Anomaly DetectiononRoad Anomaly
    AP· uses extra data· 2021-11-24
    45.1
    best: 95.21 (OodDINO)
    SOTA
    Pixel-wise Energy-biased Abstention Learning for Anomaly Segmentation on Complex Urban Driving ScenesarXiv:2111.12264
  • Anomaly DetectiononFishyscapes
    AP· uses extra data· 2021-11-24
    92.38
    best: 95.96 (RPL+CoroCL)
    SOTA
    Pixel-wise Energy-biased Abstention Learning for Anomaly Segmentation on Complex Urban Driving ScenesarXiv:2111.12264
  • Anomaly DetectiononLost and Found
    AP· uses extra data· 2021-11-24
    78.29
    best: 86.59 (Mask2Anomaly)
    SOTA
    Pixel-wise Energy-biased Abstention Learning for Anomaly Segmentation on Complex Urban Driving ScenesarXiv:2111.12264
  • Anomaly DetectiononFishyscapes L&F
    AP· uses extra data· 2021-11-24
    44.17
    best: 69.8 (cDNP+OE)
    SOTA
    Pixel-wise Energy-biased Abstention Learning for Anomaly Segmentation on Complex Urban Driving ScenesarXiv:2111.12264
  • Anomaly DetectiononRoad Anomaly
    FPR95· uses extra data· 2021-11-24
    44.58
    best: 64.69 (SynthCP)
    Pixel-wise Energy-biased Abstention Learning for Anomaly Segmentation on Complex Urban Driving ScenesarXiv:2111.12264
  • Anomaly DetectiononFishyscapes
    FPR95· uses extra data· 2021-11-24
    1.73
    best: 21.58 (FlowEneDet)
    Pixel-wise Energy-biased Abstention Learning for Anomaly Segmentation on Complex Urban Driving ScenesarXiv:2111.12264
  • Anomaly DetectiononLost and Found
    FPR· uses extra data· 2021-11-24
    0.81
    best: 44.48 (SML)
    Pixel-wise Energy-biased Abstention Learning for Anomaly Segmentation on Complex Urban Driving ScenesarXiv:2111.12264
  • Anomaly DetectiononFishyscapes L&F
    FPR95· uses extra data· 2021-11-24
    7.58
    best: 47.43 (Dirichlet DeepLab)
    Pixel-wise Energy-biased Abstention Learning for Anomaly Segmentation on Complex Urban Driving ScenesarXiv:2111.12264