Learning to Reconstruct 3D Human Pose and Shape via Model-fitting in the Loop

TL;DR

This paper introduces SPIN, a collaborative approach combining deep learning and iterative optimization for 3D human pose and shape estimation, improving accuracy and efficiency especially when 3D ground truth is limited.

Contribution

The paper proposes a novel self-improving method that integrates model-fitting with deep network training, enhancing 3D human pose estimation accuracy and speed.

Findings

Outperforms state-of-the-art methods in various settings

Effective with limited or no 3D ground truth data

Self-improving loop enhances both network and optimization results

Abstract

Model-based human pose estimation is currently approached through two different paradigms. Optimization-based methods fit a parametric body model to 2D observations in an iterative manner, leading to accurate image-model alignments, but are often slow and sensitive to the initialization. In contrast, regression-based methods, that use a deep network to directly estimate the model parameters from pixels, tend to provide reasonable, but not pixel accurate, results while requiring huge amounts of supervision. In this work, instead of investigating which approach is better, our key insight is that the two paradigms can form a strong collaboration. A reasonable, directly regressed estimate from the network can initialize the iterative optimization making the fitting faster and more accurate. Similarly, a pixel accurate fit from iterative optimization can act as strong supervision for the network. This is the core of our proposed approach SPIN (SMPL oPtimization IN the loop). The deep network initializes an iterative optimization routine that fits the body model to 2D joints within the training loop, and the fitted estimate is subsequently used to supervise the network. Our approach is self-improving by nature, since better network estimates can lead the optimization to better solutions, while more accurate optimization fits provide better supervision for the network. We demonstrate the effectiveness of our approach in different settings, where 3D ground truth is scarce, or not available, and we consistently outperform the state-of-the-art model-based pose estimation approaches by significant margins. The project website with videos, results, and code can be found at https://seas.upenn.edu/~nkolot/projects/spin.