Joint 3D Human Shape Recovery and Pose Estimation from a Single Image with Bilayer Graph

TL;DR

This paper introduces a dual-scale graph approach with fusion blocks for improved 3D human shape and pose estimation from a single image, achieving state-of-the-art results.

Contribution

It proposes a novel dual-scale graph method with fusion blocks to enhance information propagation and joint estimation of shape and pose.

Findings

Achieves state-of-the-art results on multiple datasets.

Demonstrates improved information flow between shape and pose estimation.

Outperforms previous graph-based methods in 3D human reconstruction.

Abstract

The ability to estimate the 3D human shape and pose from images can be useful in many contexts. Recent approaches have explored using graph convolutional networks and achieved promising results. The fact that the 3D shape is represented by a mesh, an undirected graph, makes graph convolutional networks a natural fit for this problem. However, graph convolutional networks have limited representation power. Information from nodes in the graph is passed to connected neighbors, and propagation of information requires successive graph convolutions. To overcome this limitation, we propose a dual-scale graph approach. We use a coarse graph, derived from a dense graph, to estimate the human's 3D pose, and the dense graph to estimate the 3D shape. Information in coarse graphs can be propagated over longer distances compared to dense graphs. In addition, information about pose can guide to recover local shape detail and vice versa. We recognize that the connection between coarse and dense is itself a graph, and introduce graph fusion blocks to exchange information between graphs with different scales. We train our model end-to-end and show that we can achieve state-of-the-art results for several evaluation datasets.