Yi Xu, Shanglin Yang, Wei Sun +3
The paper introduces a method that builds a complete 3D virtual garment model from just two RGB photos (front and back) by jointly detecting fashion landmarks and segmenting semant…