MuNet: A Mutualistic Network for Joint 3D Human Mesh Recovery and 3D Clothed Human Reconstruction from Single Images

Researchers introduced MuNet, a single-image 3D reconstruction model that jointly improves human mesh recovery and clothed human modeling using a shared graph-based framework.

MuNet uses a mutualistic design with a common 2-manifold graph representation and reciprocal training between the two tasks.

The model was evaluated on six datasets, including Human3.6M, 3DPW, MPI-INF-3DHP, THuman2.0, CAPE, and RenderPeople.

It outperformed prior methods across benchmarks, suggesting stronger accuracy and consistency for 3D human reconstruction from a single image.