- SplatVoxel: History-Aware Novel View Streaming without Temporal Training
- Quark: Real-time, High-resolution, and General Neural View Synthesis (Best Paper Award, SIGGRAPH Asia 2024)
- Text2Immersion: Generative Immersive Scene with 3D Gaussians
- Neural Free-Viewpoint Relighting for Glossy Indirect Illumination (EGSR 2023)
- Hierarchical Neural Reconstruction for Path Guiding Using Hybrid Path and Photon Samples (SIGGRAPH 2021)
Research Experience
Currently a research scientist at Google, working on Project Starline; Joined Google as a research scientist in September 2021
Education
Ph.D. in Computer Science (Google Ph.D. Fellow) from University of California, San Diego, advised by Prof. Ravi Ramamoorthi; Undergraduate in Yao Class, Tsinghua University
Background
Research interests: The combination of inverse rendering and rendering, specifically focusing on practical ways to acquire, model, and manipulate the geometry and material properties, as well as the lightings and the camera views that affect visual appearances. Works include portrait relighting, surface material modeling, hair reconstruction, path guiding, novel view synthesis, etc.