Rui Yu
I am a Staff Research Engineer at Apple, working on computer vision and machine learning. My interests include 3D reconstruction and generation, 6DoF object tracking, and synthetic data. I have contributed to Apple Device Tracking and Hand Tracking for Apple Vision Pro. My recent research includes Luce, which generates relightable 3D assets from a single image.
Bio
Before joining Apple, I worked on autonomous driving at DJI. I completed my PhD in computer vision at University College London, where my research focused on learning dense 3D models from monocular video, including reconstructing dynamic scenes and tracking deformable objects. Earlier, I studied at Northwestern Polytechnical University, where I worked on camera arrays, computational imaging, and visual tracking.
Selected Publications
-
Luce: Relightable Gaussians for 3D Asset Generation
-
Better Together: Joint Reasoning for Non-rigid 3D Reconstruction with Specularities and Shading
-
Solving Jigsaw Puzzles with Linear Programming
-
Direct, Dense, and Deformable: Template-Based Non-Rigid 3D Reconstruction from RGB Video
-
Video Pop-up: Monocular 3D Reconstruction of Dynamic Scenes
-
All-in-Focus Synthetic Aperture Imaging
-
Continuously Tracking and See-through Occlusion Based on a New Hybrid Synthetic Aperture Imaging Model
PhD thesis: Learning Dense 3D Models from Monocular Video (UCL, 2017).