W-HMR: Human Mesh Recovery in World Space with Weak-supervised Camera Calibration and Orientation Correction

29 Nov 2023  ยท  Wei Yao, Hongwen Zhang, Yunlian Sun, Jinhui Tang ยท

For a long time, in reconstructing 3D human bodies from monocular images, most methods opted to simplify the task by minimizing the influence of the camera. Using a coarse focal length setting results in the reconstructed bodies not aligning well with distorted images. Ignoring camera rotation leads to an unrealistic reconstructed body pose in world space. Consequently, the application scenarios of existing methods are confined to controlled environments. When confronted with complex and diverse in-the-wild images, they struggle to achieve accurate and reasonable reconstruction in world space. To address the above issues, we propose W-HMR, which decouples global body recovery into camera calibration, local body recovery, and global body orientation correction. We design the first weak-supervised camera calibration method for body distortion, eliminating dependence on focal length labels and achieving finer mesh-image alignment. We propose a novel orientation correction module to allow the reconstructed human body to remain normal in world space. Decoupling body orientation and body pose enables our model to consider the accuracy in camera coordinate and the reasonableness in world coordinate simultaneously, expanding the range of applications. As a result, W-HMR achieves high-quality reconstruction in dual coordinate systems, particularly in challenging scenes. Codes and demos have been released on the project page https://yw0208.github.io/w-hmr/.

PDF Abstract

Results from the Paper


 Ranked #1 on 3D Human Pose Estimation on SPEC-MTP (using extra training data)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
3D Human Pose Estimation 3DPW W-HMR PA-MPJPE 40.5 # 12
MPJPE 64.6 # 5
MPVPE 75.7 # 3
3D Human Pose Estimation AGORA W-HMR B-NMVE 70.4 # 2
B-NMJE 75.4 # 3
B-MVE 63.4 # 1
B-MPJPE 67.9 # 3
3D Human Pose Estimation Human3.6M W-HMR Average MPJPE (mm) 45.5 # 116
Multi-View or Monocular Monocular # 1
PA-MPJPE 30.2 # 7
3D Human Pose Estimation MPI-INF-3DHP W-HMR MPJPE 83.2 # 39
PA-MPJPE 59.1 # 4
3D Human Pose Estimation SPEC-MTP W-HMR W-MPJPE 118.7 # 1
W-PVE 133.9 # 1

Methods


No methods listed for this paper. Add relevant methods here