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/HINet

HINet

Reported on 64 benchmarks across 11 tasks · 2 papers · 15 SOTA

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

Computer Vision51 results

  • Image RestorationonFlare7K
    LPIPS· 2022-10-12
    0.048
    best: 0.0422 (Kotp et al)
    SOTA
    Flare7K: A Phenomenological Nighttime Flare Removal DatasetarXiv:2210.06570
  • DeblurringonGoPro
    PSNR· 2021-05-13
    32.71
    best: 35.98 (BSSTNet)
    SOTA
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Rain RemovalonTest1200
    PSNR· 2021-05-13
    33.05
    best: 34.77 (CAPTNet)
    SOTA
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Rain RemovalonTest1200
    SSIM· 2021-05-13
    0.919
    best: 0.937 (CAPTNet)
    SOTA
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Rain RemovalonTest2800
    PSNR· 2021-05-13
    33.91
    best: 34.19 (KBNet)
    SOTA
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Rain RemovalonTest2800
    SSIM· 2021-05-13
    0.941
    best: 0.944 (KBNet)
    SOTA
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Rain RemovalonTest100
    PSNR· 2021-05-13
    30.29
    best: 32 (Restormer)
    SOTA
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Rain RemovalonTest100
    SSIM· 2021-05-13
    0.906
    best: 0.923 (Restormer)
    SOTA
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • DenoisingonSIDD
    PSNR (sRGB)· 2021-05-13
    39.99
    best: 40.39 (CGNet)
    SOTA
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Image RestorationonFlare7K
    PSNR· 2022-10-12
    26.74
    best: 27.662 (Kotp et al)
    Flare7K: A Phenomenological Nighttime Flare Removal DatasetarXiv:2210.06570
  • Image RestorationonFlare7K
    SSIM· 2022-10-12
    0.882
    best: 0.901 (FF-Former)
    Flare7K: A Phenomenological Nighttime Flare Removal DatasetarXiv:2210.06570
  • Rain RemovalonRain100H
    PSNR· 2021-05-13
    30.65
    best: 34.56 (GOUB (Mean-ODE))
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Rain RemovalonRain100H
    SSIM· 2021-05-13
    0.894
    best: 0.9414 (GOUB (Mean-ODE))
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Rain RemovalonRain100L
    PSNR· 2021-05-13
    37.28
    best: 41.62 (IPT)
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Rain RemovalonRain100L
    SSIM· 2021-05-13
    0.97
    best: 0.988 (IPT)
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Image RestorationonARAD-1K
    MRAE· 2021-05-13
    0.2032
    best: 0.3814 (HSCNN+)
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Image RestorationonARAD-1K
    PSNR· 2021-05-13
    32.51
    best: 34.32 (MST++)
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Image RestorationonARAD-1K
    RMSE· 2021-05-13
    0.0303
    best: 0.0248 (MST++)
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • DenoisingonSIDD
    SSIM (sRGB)· 2021-05-13
    0.958
    best: 0.973 (NBNet)
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Instance SegmentationonA2D Sentences
    IoU mean
    0.529
    best: 0.725 (SOC (Video-Swin-B))
  • Instance SegmentationonA2D Sentences
    IoU overall
    0.679
    best: 0.807 (SOC (Video-Swin-B))
  • Instance SegmentationonA2D Sentences
    Precision@0.5
    0.611
    best: 0.851 (SOC (Video-Swin-B))
  • Instance SegmentationonA2D Sentences
    Precision@0.6
    0.559
    best: 0.827 (SOC (Video-Swin-B))
  • Instance SegmentationonA2D Sentences
    Precision@0.7
    0.486
    best: 0.767 (SgMg (Video-Swin-B))
  • Instance SegmentationonA2D Sentences
    Precision@0.8
    0.342
    best: 0.617 (SgMg (Video-Swin-B))
  • Instance SegmentationonA2D Sentences
    Precision@0.9
    0.12
    best: 0.259 (SgMg (Video-Swin-B))
  • Instance SegmentationonJ-HMDB
    IoU mean
    0.627
    best: 0.725 (SgMg (Video-Swin-B))
  • Instance SegmentationonJ-HMDB
    IoU overall
    0.652
    best: 0.737 (SgMg (Video-Swin-B))
  • Instance SegmentationonJ-HMDB
    Precision@0.5
    0.819
    best: 0.972 (SgMg (Video-Swin-B))
  • Instance SegmentationonJ-HMDB
    Precision@0.6
    0.736
    best: 0.917 (SgMg (Video-Swin-B))
  • Instance SegmentationonJ-HMDB
    Precision@0.7
    0.542
    best: 0.714 (SgMg (Video-Swin-B))
  • Instance SegmentationonJ-HMDB
    Precision@0.8
    0.168
    best: 0.225 (SgMg (Video-Swin-B))
  • Instance SegmentationonJ-HMDB
    Precision@0.9
    0.4
  • Instance SegmentationonDAVIS 2017 (val)
    J&F 1st frame
    50.2
    best: 72.5 (UNINEXT-H)
  • Instance SegmentationonDAVIS 2017 (val)
    J&F Full video
    47.9
    best: 59.4 (UniVS(Swin-L))
  • Referring Expression SegmentationonA2D Sentences
    IoU mean
    0.529
    best: 0.725 (SOC (Video-Swin-B))
  • Referring Expression SegmentationonA2D Sentences
    IoU overall
    0.679
    best: 0.807 (SOC (Video-Swin-B))
  • Referring Expression SegmentationonA2D Sentences
    Precision@0.5
    0.611
    best: 0.851 (SOC (Video-Swin-B))
  • Referring Expression SegmentationonA2D Sentences
    Precision@0.6
    0.559
    best: 0.827 (SOC (Video-Swin-B))
  • Referring Expression SegmentationonA2D Sentences
    Precision@0.7
    0.486
    best: 0.767 (SgMg (Video-Swin-B))
  • Referring Expression SegmentationonA2D Sentences
    Precision@0.8
    0.342
    best: 0.617 (SgMg (Video-Swin-B))
  • Referring Expression SegmentationonA2D Sentences
    Precision@0.9
    0.12
    best: 0.259 (SgMg (Video-Swin-B))
  • Referring Expression SegmentationonJ-HMDB
    IoU mean
    0.627
    best: 0.725 (SgMg (Video-Swin-B))
  • Referring Expression SegmentationonJ-HMDB
    IoU overall
    0.652
    best: 0.737 (SgMg (Video-Swin-B))
  • Referring Expression SegmentationonJ-HMDB
    Precision@0.5
    0.819
    best: 0.972 (SgMg (Video-Swin-B))
  • Referring Expression SegmentationonJ-HMDB
    Precision@0.6
    0.736
    best: 0.917 (SgMg (Video-Swin-B))
  • Referring Expression SegmentationonJ-HMDB
    Precision@0.7
    0.542
    best: 0.714 (SgMg (Video-Swin-B))
  • Referring Expression SegmentationonJ-HMDB
    Precision@0.8
    0.168
    best: 0.225 (SgMg (Video-Swin-B))
  • Referring Expression SegmentationonJ-HMDB
    Precision@0.9
    0.4
  • Referring Expression SegmentationonDAVIS 2017 (val)
    J&F 1st frame
    50.2
    best: 72.5 (UNINEXT-H)
  • Referring Expression SegmentationonDAVIS 2017 (val)
    J&F Full video
    47.9
    best: 59.4 (UniVS(Swin-L))

Audio7 results

  • 10-shot image generationonFlare7K
    LPIPS· 2022-10-12
    0.048
    best: 0.0422 (Kotp et al)
    SOTA
    Flare7K: A Phenomenological Nighttime Flare Removal DatasetarXiv:2210.06570
  • 10-shot image generationonGoPro
    PSNR· 2021-05-13
    32.71
    best: 35.98 (BSSTNet)
    SOTA
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • 10-shot image generationonFlare7K
    PSNR· 2022-10-12
    26.74
    best: 27.662 (Kotp et al)
    Flare7K: A Phenomenological Nighttime Flare Removal DatasetarXiv:2210.06570
  • 10-shot image generationonFlare7K
    SSIM· 2022-10-12
    0.882
    best: 0.901 (FF-Former)
    Flare7K: A Phenomenological Nighttime Flare Removal DatasetarXiv:2210.06570
  • 10-shot image generationonARAD-1K
    MRAE· 2021-05-13
    0.2032
    best: 0.3814 (HSCNN+)
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • 10-shot image generationonARAD-1K
    PSNR· 2021-05-13
    32.51
    best: 34.32 (MST++)
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • 10-shot image generationonARAD-1K
    RMSE· 2021-05-13
    0.0303
    best: 0.0248 (MST++)
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086

Medical2 results

  • Image DenoisingonSIDD
    PSNR (sRGB)· 2021-05-13
    39.99
    best: 40.39 (CGNet)
    SOTA
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • Image DenoisingonSIDD
    SSIM (sRGB)· 2021-05-13
    0.958
    best: 0.973 (NBNet)
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086

Adversarial2 results

  • 3D ArchitectureonSIDD
    PSNR (sRGB)· 2021-05-13
    39.99
    best: 40.39 (CGNet)
    SOTA
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086
  • 3D ArchitectureonSIDD
    SSIM (sRGB)· 2021-05-13
    0.958
    best: 0.973 (NBNet)
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086

Methodology1 result

  • 2D ClassificationonGoPro
    PSNR· 2021-05-13
    32.71
    best: 35.98 (BSSTNet)
    SOTA
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086

Computer Code1 result

  • Blind Image DeblurringonGoPro
    PSNR· 2021-05-13
    32.71
    best: 35.98 (BSSTNet)
    SOTA
    HINet: Half Instance Normalization Network for Image RestorationarXiv:2105.06086