Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

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

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

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

Models/ROMP

ROMP

Reported on 58 benchmarks across 8 tasks · 1 paper · 17 SOTA

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

Computer Vision33 results

Depth EstimationonRelative Human
PCDR· uses extra data· 2020-08-27
54.84
best: 68.27 (BEV)
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Depth EstimationonRelative Human
PCDR-Kid· uses extra data· 2020-08-27
48.41
best: 67.09 (BEV)
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Depth EstimationonRelative Human
mPCDK· uses extra data· 2020-08-27
0.866
best: 0.884 (BEV)
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Human Pose EstimationonEMDB
Average MPJAE (deg)· 2020-08-27
26.5975
best: 28.4555 (PyMAF)
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Human Pose EstimationonRelative Human
PCDR· uses extra data· 2020-08-27
68.27
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Pose EstimationonEMDB
Average MPJAE (deg)· 2020-08-27
26.5975
best: 28.4555 (PyMAF)
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Pose EstimationonRelative Human
PCDR· uses extra data· 2020-08-27
68.27
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Multi-Person Pose EstimationonRelative Human
PCDR· uses extra data· 2020-08-27
68.27
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Depth EstimationonRelative Human
PCDR· uses extra data· 2020-08-27
54.84
best: 68.27 (BEV)
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Depth EstimationonRelative Human
PCDR-Kid· uses extra data· 2020-08-27
48.41
best: 67.09 (BEV)
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Depth EstimationonRelative Human
mPCDK· uses extra data· 2020-08-27
0.866
best: 0.884 (BEV)
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Depth EstimationonRelative Human
PCDR-Adult· uses extra data· 2020-08-27
55.34
best: 69.71 (BEV)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Depth EstimationonRelative Human
PCDR-Baby· uses extra data· 2020-08-27
30.08
best: 60.77 (BEV)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Depth EstimationonRelative Human
PCDR-Teen· uses extra data· 2020-08-27
51.12
best: 66.07 (BEV)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Human Pose EstimationonEMDB
Average MPJAE-PA (deg)· 2020-08-27
23.9901
best: 25.7033 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Human Pose EstimationonEMDB
Average MPJPE (mm)· 2020-08-27
112.652
best: 131.065 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Human Pose EstimationonEMDB
Average MPJPE-PA (mm)· 2020-08-27
75.1869
best: 82.8502 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Human Pose EstimationonEMDB
Average MVE (mm)· 2020-08-27
134.863
best: 159.956 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Human Pose EstimationonEMDB
Average MVE-PA (mm)· 2020-08-27
90.648
best: 98.1305 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Human Pose EstimationonEMDB
Jitter (10m/s^3)· 2020-08-27
71.2556
best: 185.933 (FastMETRO-L no SMPL Head)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Human Pose Estimationon3D Poses in the Wild Challenge
MPJPE· 2020-08-27
81.76
best: 102.56 (SPIN)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Pose EstimationonEMDB
Average MPJAE-PA (deg)· 2020-08-27
23.9901
best: 25.7033 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Pose EstimationonEMDB
Average MPJPE (mm)· 2020-08-27
112.652
best: 131.065 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Pose EstimationonEMDB
Average MPJPE-PA (mm)· 2020-08-27
75.1869
best: 82.8502 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Pose EstimationonEMDB
Average MVE (mm)· 2020-08-27
134.863
best: 159.956 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Pose EstimationonEMDB
Average MVE-PA (mm)· 2020-08-27
90.648
best: 98.1305 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Pose EstimationonEMDB
Jitter (10m/s^3)· 2020-08-27
71.2556
best: 185.933 (FastMETRO-L no SMPL Head)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Pose Estimationon3D Poses in the Wild Challenge
MPJPE· 2020-08-27
81.76
best: 102.56 (SPIN)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Pose EstimationonCrowdPose
mAP @0.5:0.95· 2020-08-27
55.6
best: 83.8 (RTMO-l)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Depth EstimationonRelative Human
PCDR-Adult· uses extra data· 2020-08-27
55.34
best: 69.71 (BEV)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Depth EstimationonRelative Human
PCDR-Baby· uses extra data· 2020-08-27
30.08
best: 60.77 (BEV)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3D Depth EstimationonRelative Human
PCDR-Teen· uses extra data· 2020-08-27
51.12
best: 66.07 (BEV)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
Multi-Person Pose EstimationonCrowdPose
mAP @0.5:0.95· 2020-08-27
55.6
best: 83.8 (RTMO-l)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272

Methodology16 results

3DonEMDB
Average MPJAE (deg)· 2020-08-27
26.5975
best: 28.4555 (PyMAF)
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3DonRelative Human
PCDR· uses extra data· 2020-08-27
68.27
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3DonRelative Human
PCDR-Kid· uses extra data· 2020-08-27
48.41
best: 67.09 (BEV)
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3DonRelative Human
mPCDK· uses extra data· 2020-08-27
0.866
best: 0.884 (BEV)
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3DonEMDB
Average MPJAE-PA (deg)· 2020-08-27
23.9901
best: 25.7033 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3DonEMDB
Average MPJPE (mm)· 2020-08-27
112.652
best: 131.065 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3DonEMDB
Average MPJPE-PA (mm)· 2020-08-27
75.1869
best: 82.8502 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3DonEMDB
Average MVE (mm)· 2020-08-27
134.863
best: 159.956 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3DonEMDB
Average MVE-PA (mm)· 2020-08-27
90.648
best: 98.1305 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3DonEMDB
Jitter (10m/s^3)· 2020-08-27
71.2556
best: 185.933 (FastMETRO-L no SMPL Head)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3Don3D Poses in the Wild Challenge
MPJPE· 2020-08-27
81.76
best: 102.56 (SPIN)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3DonCrowdPose
mAP @0.5:0.95· 2020-08-27
55.6
best: 83.8 (RTMO-l)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3DonRelative Human
PCDR· uses extra data· 2020-08-27
54.84
best: 68.27
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3DonRelative Human
PCDR-Adult· uses extra data· 2020-08-27
55.34
best: 69.71 (BEV)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3DonRelative Human
PCDR-Baby· uses extra data· 2020-08-27
30.08
best: 60.77 (BEV)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
3DonRelative Human
PCDR-Teen· uses extra data· 2020-08-27
51.12
best: 66.07 (BEV)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272

Audio10 results

1 Image, 2*2 StitchionEMDB
Average MPJAE (deg)· 2020-08-27
26.5975
best: 28.4555 (PyMAF)
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
1 Image, 2*2 StitchionRelative Human
PCDR· uses extra data· 2020-08-27
68.27
SOTA
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
1 Image, 2*2 StitchionEMDB
Average MPJAE-PA (deg)· 2020-08-27
23.9901
best: 25.7033 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
1 Image, 2*2 StitchionEMDB
Average MPJPE (mm)· 2020-08-27
112.652
best: 131.065 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
1 Image, 2*2 StitchionEMDB
Average MPJPE-PA (mm)· 2020-08-27
75.1869
best: 82.8502 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
1 Image, 2*2 StitchionEMDB
Average MVE (mm)· 2020-08-27
134.863
best: 159.956 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
1 Image, 2*2 StitchionEMDB
Average MVE-PA (mm)· 2020-08-27
90.648
best: 98.1305 (PyMAF)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
1 Image, 2*2 StitchionEMDB
Jitter (10m/s^3)· 2020-08-27
71.2556
best: 185.933 (FastMETRO-L no SMPL Head)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
1 Image, 2*2 Stitchion3D Poses in the Wild Challenge
MPJPE· 2020-08-27
81.76
best: 102.56 (SPIN)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272
1 Image, 2*2 StitchionCrowdPose
mAP @0.5:0.95· 2020-08-27
55.6
best: 83.8 (RTMO-l)
Monocular, One-stage, Regression of Multiple 3D People arXiv:2008.12272