3D Human Pose Estimation
3D Human Pose Estimation is a computer vision task that involves estimating the 3D positions and orientations of body joints and bones from 2D images or videos. The goal is to reconstruct the 3D pose of a person in real-time, which can be used in a variety of applications, such as virtual reality, human-computer interaction, and motion analysis.
Papers
Showing 1–25 of 665 papers
All datasets3DPWMPI-INF-3DHPHuman3.6MHumanEva-IH3WBTotal CaptureEMDBAGORAPanopticSurreal3D Poses in the Wild ChallengeAIST++
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Simple-baseline | PA-MPJPE | 157 | — | Unverified |
| 2 | HMR | MPJPE | 130 | — | Unverified |
| 3 | BMP | MPVPE | 119.3 | — | Unverified |
| 4 | SPIN | MPVPE | 116.4 | — | Unverified |
| 5 | Wenshuo et a;. | MPVPE | 112.6 | — | Unverified |
| 6 | TCMR (T=16 w/o 3DPW) | MPVPE | 111.5 | — | Unverified |
| 7 | CHOMP | MPVPE | 110.1 | — | Unverified |
| 8 | PC-HMR | MPVPE | 108.6 | — | Unverified |
| 9 | 3DCrowdNet | MPVPE | 108.5 | — | Unverified |
| 10 | SMPLify | PA-MPJPE | 106.8 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | VNect (Augm.) | MPJPE | 124.7 | — | Unverified |
| 2 | HMR | MPJPE | 124.2 | — | Unverified |
| 3 | Single-Shot Multi-Person | MPJPE | 122.2 | — | Unverified |
| 4 | Mehta | MPJPE | 117.6 | — | Unverified |
| 5 | PONet | MPJPE | 115 | — | Unverified |
| 6 | Pose Consensus (monocular) | MPJPE | 112.1 | — | Unverified |
| 7 | GeoRep (fully-supervised) | MPJPE | 110.8 | — | Unverified |
| 8 | XFormer (HRNet) | MPJPE | 109.8 | — | Unverified |
| 9 | EpipolarPose (fully-supervised) | MPJPE | 108.99 | — | Unverified |
| 10 | SPIN | MPJPE | 105.2 | — | Unverified |