https://ieeexplore.ieee.org/document/11414185

http://english.cas.cn/newsroom/research-news/202603/t20260303_1151362.shtml

A research team led by Prof. SONG Zhan from the CAS Shenzhen Institute of Advanced Technology, together with Dr. YE Yuping from Fujian University of Technology, has developed a new curvature‑fused graph attention network (CF‑GAT)capable of predicting facial landmarks directly from raw point clouds.

Based on this tool, they developed a custom 3D/4D facial acquisition system and conducted standardized data collection with approximately 200,000 high-fidelity 3D facial scans, complemented by a multi-expression 3D face dataset, a standardized 3D facial landmark dataset, a high-precision 3D human body dataset, and a dynamic 4D facial expression dataset. This multimodal biometric dataset was selected for Fujian Province’s 2025 High-Quality AI Dataset Program.

Based on such studies, humanoid robots will become more and indistinguishable from real human, enabling virtual humans to express vivid emotions, recognize identities, and demonstrate embodied intelligence.