YuliangXiu/ECON
CVPR 2023 paper implementing a neural network-based system for reconstructing 3D clothed human models from single images.

ECON (Explicit Clothed humans Optimized via Normal integration) is a computer vision research project that reconstructs full 3D human meshes with clothing from monocular images. It leverages deep learning with PyTorch, builds on SMPL/SMPLX body models, and uses pixel implicit function (PIFu) approaches combined with normal map integration to produce detailed avatar geometry suitable for virtual reality, digital twins, and metaverse applications.