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/PAN

PAN

Reported on 13 benchmarks across 2 tasks · 4 papers · 8 SOTA

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

Computer Vision16 results

  • Scene Text DetectiononMSRA-TD500
    F-Measure· 2019-08-16
    84.1
    best: 89.4 (MixNet)
    SOTA
    Efficient and Accurate Arbitrary-Shaped Text Detection with Pixel Aggregation NetworkarXiv:1908.05900
  • Scene Text DetectiononMSRA-TD500
    Recall· 2019-08-16
    83.8
    best: 88.1 (MixNet)
    SOTA
    Efficient and Accurate Arbitrary-Shaped Text Detection with Pixel Aggregation NetworkarXiv:1908.05900
  • Scene Text DetectiononSCUT-CTW1500
    F-Measure· 2018-11-22
    85
    best: 89.8 (MixNet)
    SOTA
    Mask R-CNN with Pyramid Attention Network for Scene Text DetectionarXiv:1811.09058
  • Scene Text DetectiononSCUT-CTW1500
    FPS· 2018-11-22
    65.2
    best: 129.1 (FAST-T-512)
    SOTA
    Mask R-CNN with Pyramid Attention Network for Scene Text DetectionarXiv:1811.09058
  • Scene Text DetectiononSCUT-CTW1500
    Precision· 2018-11-22
    86.8
    best: 91.7 (DPText-DETR (ResNet50))
    SOTA
    Mask R-CNN with Pyramid Attention Network for Scene Text DetectionarXiv:1811.09058
  • Scene Text DetectiononICDAR 2015
    F-Measure· 2017-04-11
    82.9
    best: 92.23 (TextFuseNet (ResNeXt-101))
    SOTA
    EAST: An Efficient and Accurate Scene Text DetectorarXiv:1704.03155
  • Scene Text DetectiononICDAR 2015
    Precision· 2017-04-11
    84
    best: 93.96 (TextFuseNet (ResNeXt-101))
    SOTA
    EAST: An Efficient and Accurate Scene Text DetectorarXiv:1704.03155
  • Scene Text DetectiononICDAR 2015
    Recall· 2017-04-11
    81.9
    best: 91.98 (CharNet H-88 (single-scale))
    SOTA
    EAST: An Efficient and Accurate Scene Text DetectorarXiv:1704.03155
  • Scene Text DetectiononSCUT-CTW1500
    Recall· 2018-11-22
    83.2
    best: 88.3 (MixNet)
    Mask R-CNN with Pyramid Attention Network for Scene Text DetectionarXiv:1811.09058
  • Scene Text DetectiononICDAR 2017 MLT
    Precision· 2018-11-22
    80
    best: 84.42 (PMTD*)
    Mask R-CNN with Pyramid Attention Network for Scene Text DetectionarXiv:1811.09058
  • Scene Text DetectiononICDAR 2017 MLT
    Recall· 2018-11-22
    69.8
    best: 76.44 (SBD)
    Mask R-CNN with Pyramid Attention Network for Scene Text DetectionarXiv:1811.09058
  • Scene Text DetectiononICDAR 2015
    F-Measure· 2018-11-22
    85.9
    best: 92.23 (TextFuseNet (ResNeXt-101))
    Mask R-CNN with Pyramid Attention Network for Scene Text DetectionarXiv:1811.09058
  • Scene Text DetectiononICDAR 2015
    Precision· 2018-11-22
    90.8
    best: 93.96 (TextFuseNet (ResNeXt-101))
    Mask R-CNN with Pyramid Attention Network for Scene Text DetectionarXiv:1811.09058
  • Scene Text DetectiononICDAR 2015
    Recall· 2018-11-22
    81.5
    best: 91.98 (CharNet H-88 (single-scale))
    Mask R-CNN with Pyramid Attention Network for Scene Text DetectionarXiv:1811.09058
  • Person Re-IdentificationonDukeMTMC-reID
    Rank-1· 2017-07-03
    71.59
    best: 95.6 (CTL Model (ResNet50, 256x128))
    Pedestrian Alignment Network for Large-scale Person Re-identificationarXiv:1707.00408
  • Person Re-IdentificationonDukeMTMC-reID
    mAP· 2017-07-03
    51.51
    best: 97.1 (DenseIL)
    Pedestrian Alignment Network for Large-scale Person Re-identificationarXiv:1707.00408