Lixiang Lin, Jianke Zhu: Semantic-preserved point-based human avatar. Comput. Vis. Image Underst. 252: 104307 (2025)