VirtualPose: Learning Generalizable 3D Human Pose Models from Virtual Data
This addresses the generalization problem in 3D human pose estimation for practical applications, though it is incremental as it builds on existing methods with a novel training approach.
The paper tackles the poor generalization of monocular 3D human pose estimation methods across different cameras, poses, and appearances by introducing VirtualPose, a two-stage framework that uses virtual data to train models, achieving state-of-the-art results without relying on paired image-3D pose benchmarks.
While monocular 3D pose estimation seems to have achieved very accurate results on the public datasets, their generalization ability is largely overlooked. In this work, we perform a systematic evaluation of the existing methods and find that they get notably larger errors when tested on different cameras, human poses and appearance. To address the problem, we introduce VirtualPose, a two-stage learning framework to exploit the hidden "free lunch" specific to this task, i.e. generating infinite number of poses and cameras for training models at no cost. To that end, the first stage transforms images to abstract geometry representations (AGR), and then the second maps them to 3D poses. It addresses the generalization issue from two aspects: (1) the first stage can be trained on diverse 2D datasets to reduce the risk of over-fitting to limited appearance; (2) the second stage can be trained on diverse AGR synthesized from a large number of virtual cameras and poses. It outperforms the SOTA methods without using any paired images and 3D poses from the benchmarks, which paves the way for practical applications. Code is available at https://github.com/wkom/VirtualPose.