3D Human Pose and Shape Estimation via HybrIK-Transformer
This work addresses the problem of more efficient and accurate 3D human pose estimation for computer vision applications, but it is incremental as it builds directly on the existing HybrIK framework.
The paper tackles 3D human pose and shape estimation from 2D monocular images by enhancing the HybrIK method, replacing its deconvolution module with a Transformer to improve accuracy and computational efficiency, as demonstrated on datasets like H36M, PW3D, COCO, and HP3D.
HybrIK relies on a combination of analytical inverse kinematics and deep learning to produce more accurate 3D pose estimation from 2D monocular images. HybrIK has three major components: (1) pretrained convolution backbone, (2) deconvolution to lift 3D pose from 2D convolution features, (3) analytical inverse kinematics pass correcting deep learning prediction using learned distribution of plausible twist and swing angles. In this paper we propose an enhancement of the 2D to 3D lifting module, replacing deconvolution with Transformer, resulting in accuracy and computational efficiency improvement relative to the original HybrIK method. We demonstrate our results on commonly used H36M, PW3D, COCO and HP3D datasets. Our code is publicly available https://github.com/boreshkinai/hybrik-transformer.