Scholar
Hongyu Xu
Google Scholar ID: x0onCYgAAAAJ
Research Scientist, Meta Reality Labs
Citations & Impact
Citations
847
H-index
11
i10-index
11
Publications
20
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- Nov, 2025: Best Paper Award from ACM MM 2025 Multimodal Foundation Models for Spatial Intelligence Workshop; DynamicVerse: Physically-Aware Multimodal Modeling for Dynamic 4D Worlds accepted to NeurIPS 2025; VLM-3R project online; MV-DUSt3R+ accepted as an Oral at CVPR 2025; MV-DUSt3R+ open sourced; Dec, 2024: MV-DUSt3R+ online; Jun, 2024: Room Tracking on VisionPro unveiled at Apple WWDC 2024; Oct, 2023: Paper “RoomDreamer: Text-Driven 3D Indoor Scene Synthesis with Coherent Geometry and Texture” accepted to ACM Multimedia 2023; Jun, 2023: RoomPlan Enhancement introduced at Apple WWDC 2023; Oct, 2022: Research article “3D Parametric Room Representation with RoomPlan” published at Apple Machine Learning Research; Jun, 2022: RoomPlan first released at Apple WWDC 2022.
Research Experience
- Before joining Meta Reality Labs, he was a technical lead and senior machine learning/computer vision engineer with the Video Engineering Group at Apple Inc., leading the algorithm development and delivery of multiple groundbreaking products, including Room Tracking on VisionPro, RoomPlan Enhancement, and RoomPlan. Additionally, he collaborated with Apple AIML on 3D Scene Style Generation, pioneering RoomDreamer.
Education
- Ph.D. and M.S. from University of Maryland, College Park, advised by Prof. Rama Chellappa; B.S. in Electrical Engineering and Information Science from University of Science and Technology of China.
Background
- Research Interests: 3D Vision-Language Models, Generative AI, 3D Reconstruction, Spatial Perception. Background: Research Scientist at Meta Reality Labs, focusing on developing advanced on-device solutions to strengthen the perception stack for Meta’s MR/VR product lines.
Miscellany
- Currently residing in Burlingame, California; working at Meta Inc.; personal interests not mentioned.
Co-authors