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/SHOT

SHOT

Reported on 106 benchmarks across 8 tasks · 1 paper · 4 SOTA

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

Methodology73 results

  • Domain AdaptationonUSPS-to-MNIST
    Accuracy· 2020-02-20
    98.4
    best: 98.75 (FAMCD)
    SOTA
    Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain AdaptationarXiv:2002.08546
  • Domain AdaptationonVisDA-2017
    Accuracy· 2020-02-20
    82.9
    best: 93.2 (RCL)
    SOTA
    Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain AdaptationarXiv:2002.08546
  • Domain AdaptationonOffice-Home
    Accuracy (%)· 2020-02-20
    78.3
    best: 80.2 (EvoADA)
    SOTA
    Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain AdaptationarXiv:2002.08546
  • Domain AdaptationonSVHN-to-MNIST
    Accuracy· 2020-02-20
    98.9
    best: 99.18 (Mean teacher)
    Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain AdaptationarXiv:2002.08546
  • Domain AdaptationonOffice-31
    Average Accuracy· 2020-02-20
    88.6
    best: 96 (FFTAT)
    Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain AdaptationarXiv:2002.08546
  • Domain AdaptationonSVNH-to-MNIST
    Accuracy· 2020-02-20
    98.9
    best: 98.91 (SRDA (RAN))
    Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain AdaptationarXiv:2002.08546
  • Domain AdaptationonVisDA2017
    Accuracy· 2020-02-20
    82.9
    best: 93.8 (FFTAT)
    Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain AdaptationarXiv:2002.08546
  • Domain AdaptationonMNIST-to-USPS
    Accuracy· 2020-02-20
    98
    best: 98.8 (FACT)
    Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain AdaptationarXiv:2002.08546
  • Domain AdaptationonOffice-Home
    Accuracy
    71.8
    best: 92.3 (SWG)
  • 3DonWildtrack
    MODA
    90.2
    best: 94.1 (MVFP)
  • 3DonWildtrack
    MODP
    76.5
    best: 82.1 (MVDeTr)
  • 3DonWildtrack
    Recall
    94
    best: 97.7 (MVFP)
  • 3DonCityStreet
    F1_score (2m)
    71.8
    best: 79.2 (3DROM)
  • 3DonCityStreet
    MODA (2m)
    53.5
    best: 60 (3DROM)
  • 3DonCityStreet
    MODP (2m)
    72.4
    best: 74.1 (MVDeTr)
  • 3DonCityStreet
    Precision (2m)
    91
    best: 92.8 (MVDeTr)
  • 3DonCityStreet
    Recall (2m)
    59.4
    best: 76.2 (3DROM)
  • 3DonCVCS
    F1_score (1m)
    67
    best: 68.4 (SVCW)
  • 3DonCVCS
    MODA (1m)
    45
    best: 46.2 (SVCW)
  • 3DonCVCS
    MODP (1m)
    77.4
    best: 84.1 (MVDeTr)
  • 3DonCVCS
    Precision (1m)
    83.6
    best: 95.3 (MVDeTr)
  • 3DonCVCS
    Recall (1m)
    55.9
    best: 59.1 (SVCW)
  • 3DonMultiviewX
    MODA
    88.3
    best: 96.7 (M-MVOT)
  • 3DonMultiviewX
    MODP
    82
    best: 91.3 (MVDeTr)
  • 3DonMultiviewX
    Recall
    91.5
    best: 97.9 (M-MVOT)
  • 2D ClassificationonWildtrack
    MODA
    90.2
    best: 94.1 (MVFP)
  • 2D ClassificationonWildtrack
    MODP
    76.5
    best: 82.1 (MVDeTr)
  • 2D ClassificationonWildtrack
    Recall
    94
    best: 97.7 (MVFP)
  • 2D ClassificationonCityStreet
    F1_score (2m)
    71.8
    best: 79.2 (3DROM)
  • 2D ClassificationonCityStreet
    MODA (2m)
    53.5
    best: 60 (3DROM)
  • 2D ClassificationonCityStreet
    MODP (2m)
    72.4
    best: 74.1 (MVDeTr)
  • 2D ClassificationonCityStreet
    Precision (2m)
    91
    best: 92.8 (MVDeTr)
  • 2D ClassificationonCityStreet
    Recall (2m)
    59.4
    best: 76.2 (3DROM)
  • 2D ClassificationonCVCS
    F1_score (1m)
    67
    best: 68.4 (SVCW)
  • 2D ClassificationonCVCS
    MODA (1m)
    45
    best: 46.2 (SVCW)
  • 2D ClassificationonCVCS
    MODP (1m)
    77.4
    best: 84.1 (MVDeTr)
  • 2D ClassificationonCVCS
    Precision (1m)
    83.6
    best: 95.3 (MVDeTr)
  • 2D ClassificationonCVCS
    Recall (1m)
    55.9
    best: 59.1 (SVCW)
  • 2D ClassificationonMultiviewX
    MODA
    88.3
    best: 96.7 (M-MVOT)
  • 2D ClassificationonMultiviewX
    MODP
    82
    best: 91.3 (MVDeTr)
  • 2D ClassificationonMultiviewX
    Recall
    91.5
    best: 97.9 (M-MVOT)
  • 2D Object DetectiononWildtrack
    MODA
    90.2
    best: 94.1 (MVFP)
  • 2D Object DetectiononWildtrack
    MODP
    76.5
    best: 82.1 (MVDeTr)
  • 2D Object DetectiononWildtrack
    Recall
    94
    best: 97.7 (MVFP)
  • 2D Object DetectiononCityStreet
    F1_score (2m)
    71.8
    best: 79.2 (3DROM)
  • 2D Object DetectiononCityStreet
    MODA (2m)
    53.5
    best: 60 (3DROM)
  • 2D Object DetectiononCityStreet
    MODP (2m)
    72.4
    best: 74.1 (MVDeTr)
  • 2D Object DetectiononCityStreet
    Precision (2m)
    91
    best: 92.8 (MVDeTr)
  • 2D Object DetectiononCityStreet
    Recall (2m)
    59.4
    best: 76.2 (3DROM)
  • 2D Object DetectiononCVCS
    F1_score (1m)
    67
    best: 68.4 (SVCW)
  • 2D Object DetectiononCVCS
    MODA (1m)
    45
    best: 46.2 (SVCW)
  • 2D Object DetectiononCVCS
    MODP (1m)
    77.4
    best: 84.1 (MVDeTr)
  • 2D Object DetectiononCVCS
    Precision (1m)
    83.6
    best: 95.3 (MVDeTr)
  • 2D Object DetectiononCVCS
    Recall (1m)
    55.9
    best: 59.1 (SVCW)
  • 2D Object DetectiononMultiviewX
    MODA
    88.3
    best: 96.7 (M-MVOT)
  • 2D Object DetectiononMultiviewX
    MODP
    82
    best: 91.3 (MVDeTr)
  • 2D Object DetectiononMultiviewX
    Recall
    91.5
    best: 97.9 (M-MVOT)
  • 16konWildtrack
    MODA
    90.2
    best: 94.1 (MVFP)
  • 16konWildtrack
    MODP
    76.5
    best: 82.1 (MVDeTr)
  • 16konWildtrack
    Recall
    94
    best: 97.7 (MVFP)
  • 16konCityStreet
    F1_score (2m)
    71.8
    best: 79.2 (3DROM)
  • 16konCityStreet
    MODA (2m)
    53.5
    best: 60 (3DROM)
  • 16konCityStreet
    MODP (2m)
    72.4
    best: 74.1 (MVDeTr)
  • 16konCityStreet
    Precision (2m)
    91
    best: 92.8 (MVDeTr)
  • 16konCityStreet
    Recall (2m)
    59.4
    best: 76.2 (3DROM)
  • 16konCVCS
    F1_score (1m)
    67
    best: 68.4 (SVCW)
  • 16konCVCS
    MODA (1m)
    45
    best: 46.2 (SVCW)
  • 16konCVCS
    MODP (1m)
    77.4
    best: 84.1 (MVDeTr)
  • 16konCVCS
    Precision (1m)
    83.6
    best: 95.3 (MVDeTr)
  • 16konCVCS
    Recall (1m)
    55.9
    best: 59.1 (SVCW)
  • 16konMultiviewX
    MODA
    88.3
    best: 96.7 (M-MVOT)
  • 16konMultiviewX
    MODP
    82
    best: 91.3 (MVDeTr)
  • 16konMultiviewX
    Recall
    91.5
    best: 97.9 (M-MVOT)

Computer Vision33 results

  • Source-Free Domain AdaptationonVisDA-2017
    Accuracy· 2020-02-20
    82.9
    best: 93.2 (RCL)
    SOTA
    Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain AdaptationarXiv:2002.08546
  • Object DetectiononWildtrack
    MODA
    90.2
    best: 94.1 (MVFP)
  • Object DetectiononWildtrack
    MODP
    76.5
    best: 82.1 (MVDeTr)
  • Object DetectiononWildtrack
    Recall
    94
    best: 97.7 (MVFP)
  • Object DetectiononCityStreet
    F1_score (2m)
    71.8
    best: 79.2 (3DROM)
  • Object DetectiononCityStreet
    MODA (2m)
    53.5
    best: 60 (3DROM)
  • Object DetectiononCityStreet
    MODP (2m)
    72.4
    best: 74.1 (MVDeTr)
  • Object DetectiononCityStreet
    Precision (2m)
    91
    best: 92.8 (MVDeTr)
  • Object DetectiononCityStreet
    Recall (2m)
    59.4
    best: 76.2 (3DROM)
  • Object DetectiononCVCS
    F1_score (1m)
    67
    best: 68.4 (SVCW)
  • Object DetectiononCVCS
    MODA (1m)
    45
    best: 46.2 (SVCW)
  • Object DetectiononCVCS
    MODP (1m)
    77.4
    best: 84.1 (MVDeTr)
  • Object DetectiononCVCS
    Precision (1m)
    83.6
    best: 95.3 (MVDeTr)
  • Object DetectiononCVCS
    Recall (1m)
    55.9
    best: 59.1 (SVCW)
  • Object DetectiononMultiviewX
    MODA
    88.3
    best: 96.7 (M-MVOT)
  • Object DetectiononMultiviewX
    MODP
    82
    best: 91.3 (MVDeTr)
  • Object DetectiononMultiviewX
    Recall
    91.5
    best: 97.9 (M-MVOT)
  • 3D Object DetectiononWildtrack
    MODA
    90.2
    best: 94.1 (MVFP)
  • 3D Object DetectiononWildtrack
    MODP
    76.5
    best: 82.1 (MVDeTr)
  • 3D Object DetectiononWildtrack
    Recall
    94
    best: 97.7 (MVFP)
  • 3D Object DetectiononCityStreet
    F1_score (2m)
    71.8
    best: 79.2 (3DROM)
  • 3D Object DetectiononCityStreet
    MODA (2m)
    53.5
    best: 60 (3DROM)
  • 3D Object DetectiononCityStreet
    MODP (2m)
    72.4
    best: 74.1 (MVDeTr)
  • 3D Object DetectiononCityStreet
    Precision (2m)
    91
    best: 92.8 (MVDeTr)
  • 3D Object DetectiononCityStreet
    Recall (2m)
    59.4
    best: 76.2 (3DROM)
  • 3D Object DetectiononCVCS
    F1_score (1m)
    67
    best: 68.4 (SVCW)
  • 3D Object DetectiononCVCS
    MODA (1m)
    45
    best: 46.2 (SVCW)
  • 3D Object DetectiononCVCS
    MODP (1m)
    77.4
    best: 84.1 (MVDeTr)
  • 3D Object DetectiononCVCS
    Precision (1m)
    83.6
    best: 95.3 (MVDeTr)
  • 3D Object DetectiononCVCS
    Recall (1m)
    55.9
    best: 59.1 (SVCW)
  • 3D Object DetectiononMultiviewX
    MODA
    88.3
    best: 96.7 (M-MVOT)
  • 3D Object DetectiononMultiviewX
    MODP
    82
    best: 91.3 (MVDeTr)
  • 3D Object DetectiononMultiviewX
    Recall
    91.5
    best: 97.9 (M-MVOT)