We also introduced a better VAE-based pose prior, VPoser, trained on AMASS, and we improved the interpenetration detection.īecause images with ground-truth human pose and shape are hard to obtain, these optimization methods provide critical pseudo ground truth for training deep regression networks. SMPLify-X introduced several improvements including a gender classifier so that the estimated body shapes better matched the image.
With SMPLify-X we extend this concept to estimate the expressive SMPL-X model by fitting it to 2D landmarks from OpenPose. Because of the inherent ambiguity in estimating 3D from 2D, SMPLify introduced a pose prior trained on mocap data and a term that discouraged self-penetration. We introduced the first such method, SMPLify, which optimizes SMPL pose and shape to minimize the 2D error between detected joints and projected SMPL joints. Optimization-based approaches directly fit a 3D body model like SMPL to image observations (e.g., detected joint locations, edges, silhouettes, semantic segmentations, etc.). In our view, the two approaches are not competing, but rather, complimentary. While typically slower than regression methods, optimization approaches require no training data, can be quickly adapted to new problems, and produce image-aligned results. While data-driven methods for directly regressing 3D humans from 2D images are widely popular, optimization-based methods continue to play an important role.
Theory of Inhomogeneous Condensed Matter.