Domain Adaptive 3D Pose Augmentation for In-the-wild Human Mesh Recovery

TL;DR

This paper introduces DAPA, a domain adaptive data augmentation technique that improves 3D human mesh recovery models' ability to generalize to real-world scenarios by combining synthetic supervision and domain adaptation.

Contribution

It proposes a novel augmentation method, DAPA, that enhances model generalization to in-the-wild data by integrating synthetic mesh supervision with 2D keypoint domain adaptation.

Findings

DAPA improves performance on 3DPW and AGORA benchmarks.

Finetuning with DAPA yields significant accuracy gains.

Demonstrated effectiveness on real-world parent-child interaction videos.

Abstract

The ability to perceive 3D human bodies from a single image has a multitude of applications ranging from entertainment and robotics to neuroscience and healthcare. A fundamental challenge in human mesh recovery is in collecting the ground truth 3D mesh targets required for training, which requires burdensome motion capturing systems and is often limited to indoor laboratories. As a result, while progress is made on benchmark datasets collected in these restrictive settings, models fail to generalize to real-world "in-the-wild" scenarios due to distribution shifts. We propose Domain Adaptive 3D Pose Augmentation (DAPA), a data augmentation method that enhances the model's generalization ability in in-the-wild scenarios. DAPA combines the strength of methods based on synthetic datasets by getting direct supervision from the synthesized meshes, and domain adaptation methods by using ground truth 2D keypoints from the target dataset. We show quantitatively that finetuning with DAPA effectively improves results on benchmarks 3DPW and AGORA. We further demonstrate the utility of DAPA on a challenging dataset curated from videos of real-world parent-child interaction.