Deep unsupervised 3D human body reconstruction from a sparse set of landmarks
arXiv:2106.12282 · doi:10.1007/s11263-021-01488-2
Abstract
In this paper we propose the first deep unsupervised approach in human body reconstruction to estimate body surface from a sparse set of landmarks, so called DeepMurf. We apply a denoising autoencoder to estimate missing landmarks. Then we apply an attention model to estimate body joints from landmarks. Finally, a cascading network is applied to regress parameters of a statistical generative model that reconstructs body. Our set of proposed loss functions allows us to train the network in an unsupervised way. Results on four public datasets show that our approach accurately reconstructs the human body from real world mocap data.
References in corpus (7)
- Point-Set Registration: Coherent Point Drift
- Learning from Synthetic Humans
- XNect: Real-time Multi-Person 3D Motion Capture with a Single RGB Camera
- Unsupervised Adversarial Learning of 3D Human Pose from 2D Joint Locations
- SMPLR: Deep SMPL reverse for 3D human pose and shape recovery
- LoopReg: Self-supervised Learning of Implicit Surface Correspondences, Pose and Shape for 3D Human Mesh Registration
- Unsupervised Learning of Shape and Pose with Differentiable Point Clouds