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/DINOv2 (ViT-g/14 frozen, w/ DPT decoder)

DINOv2 (ViT-g/14 frozen, w/ DPT decoder)

Reported on 28 benchmarks across 2 tasks · 1 paper · 4 SOTA

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

Computer Vision14 results

  • Depth EstimationonNYU-Depth V2
    Delta < 1.25^2· uses extra data· 2023-04-14
    0.996
    best: 1 (HybridDepth)
    SOTA
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • Depth EstimationonNYU-Depth V2
    Delta < 1.25^3· uses extra data· 2023-04-14
    0.9994
    best: 1 (HybridDepth)
    SOTA
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • Depth EstimationonNYU-Depth V2
    RMS· uses extra data· 2023-04-14
    0.279
    best: 0.792 (PAD-Net)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • Depth EstimationonNYU-Depth V2
    Delta < 1.25· uses extra data· 2023-04-14
    0.9497
    best: 0.989 (UniK3D (FT, metric))
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • Depth EstimationonNYU-Depth V2
    RMSE· uses extra data· 2023-04-14
    0.279
    best: 0.013 (Defocus/DepthNet (Normalized))
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • Depth EstimationonNYU-Depth V2
    absolute relative error· uses extra data· 2023-04-14
    0.0907
    best: 0.026 (HybridDepth)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • Depth EstimationonNYU-Depth V2
    log 10· uses extra data· 2023-04-14
    0.0371
    best: 0.059 (SC-DepthV2)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • Depth EstimationonKITTI Eigen split
    Delta < 1.25· 2023-04-14
    0.968
    best: 0.99 (SPIDepth)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • Depth EstimationonKITTI Eigen split
    Delta < 1.25^2· 2023-04-14
    0.997
    best: 0.999 (SPIDepth)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • Depth EstimationonKITTI Eigen split
    Delta < 1.25^3· 2023-04-14
    0.9993
    best: 1 (SPIDepth)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • Depth EstimationonKITTI Eigen split
    RMSE· 2023-04-14
    2.1128
    best: 1.394 (SPIDepth)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • Depth EstimationonKITTI Eigen split
    RMSE log· 2023-04-14
    0.0882
    best: 0.048 (SPIDepth)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • Depth EstimationonKITTI Eigen split
    Sq Rel· 2023-04-14
    0.1797
    best: 0.224 (SfM-Revisited)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • Depth EstimationonKITTI Eigen split
    absolute relative error· 2023-04-14
    0.0652
    best: 0.029 (SPIDepth)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193

Methodology14 results

  • 3DonNYU-Depth V2
    Delta < 1.25^2· uses extra data· 2023-04-14
    0.996
    best: 1 (HybridDepth)
    SOTA
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • 3DonNYU-Depth V2
    Delta < 1.25^3· uses extra data· 2023-04-14
    0.9994
    best: 1 (HybridDepth)
    SOTA
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • 3DonNYU-Depth V2
    RMS· uses extra data· 2023-04-14
    0.279
    best: 0.792 (PAD-Net)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • 3DonNYU-Depth V2
    Delta < 1.25· uses extra data· 2023-04-14
    0.9497
    best: 0.989 (UniK3D (FT, metric))
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • 3DonNYU-Depth V2
    RMSE· uses extra data· 2023-04-14
    0.279
    best: 0.013 (Defocus/DepthNet (Normalized))
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • 3DonNYU-Depth V2
    absolute relative error· uses extra data· 2023-04-14
    0.0907
    best: 0.026 (HybridDepth)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • 3DonNYU-Depth V2
    log 10· uses extra data· 2023-04-14
    0.0371
    best: 0.059 (SC-DepthV2)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • 3DonKITTI Eigen split
    Delta < 1.25· 2023-04-14
    0.968
    best: 0.99 (SPIDepth)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • 3DonKITTI Eigen split
    Delta < 1.25^2· 2023-04-14
    0.997
    best: 0.999 (SPIDepth)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • 3DonKITTI Eigen split
    Delta < 1.25^3· 2023-04-14
    0.9993
    best: 1 (SPIDepth)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • 3DonKITTI Eigen split
    RMSE· 2023-04-14
    2.1128
    best: 1.394 (SPIDepth)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • 3DonKITTI Eigen split
    RMSE log· 2023-04-14
    0.0882
    best: 0.048 (SPIDepth)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • 3DonKITTI Eigen split
    Sq Rel· 2023-04-14
    0.1797
    best: 0.224 (SfM-Revisited)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193
  • 3DonKITTI Eigen split
    absolute relative error· 2023-04-14
    0.0652
    best: 0.029 (SPIDepth)
    DINOv2: Learning Robust Visual Features without SupervisionarXiv:2304.07193