Image-based Synthesis for Deep 3D Human Pose Estimation

TL;DR

This paper presents a novel image-based synthesis engine that generates photorealistic synthetic images with 3D human pose annotations, significantly enhancing training data for CNN-based 3D pose estimation in diverse environments.

Contribution

The authors introduce a new image synthesis method that combines real images and 3D motion data to create large, annotated datasets without domain adaptation, improving 3D human pose estimation.

Findings

Outperforms existing methods on Human3.6M dataset

Shows promising results on real-world images (LSP)

Synthetic images generalize well to real images

Abstract

This paper addresses the problem of 3D human pose estimation in the wild. A significant challenge is the lack of training data, i.e., 2D images of humans annotated with 3D poses. Such data is necessary to train state-of-the-art CNN architectures. Here, we propose a solution to generate a large set of photorealistic synthetic images of humans with 3D pose annotations. We introduce an image-based synthesis engine that artificially augments a dataset of real images with 2D human pose annotations using 3D motion capture data. Given a candidate 3D pose, our algorithm selects for each joint an image whose 2D pose locally matches the projected 3D pose. The selected images are then combined to generate a new synthetic image by stitching local image patches in a kinematically constrained manner. The resulting images are used to train an end-to-end CNN for full-body 3D pose estimation. We cluster the training data into a large number of pose classes and tackle pose estimation as a $K$-way classification problem. Such an approach is viable only with large training sets such as ours. Our method outperforms most of the published works in terms of 3D pose estimation in controlled environments (Human3.6M) and shows promising results for real-world images (LSP). This demonstrates that CNNs trained on artificial images generalize well to real images. Compared to data generated from more classical rendering engines, our synthetic images do not require any domain adaptation or fine-tuning stage.