
I am a Research Scientist and Manager at ByteDance AI Lab on computer graphics and 3D computer vision. Our research and development support the ever-growing product line of ByteDance, including TikTok (Douyin), Toutiao and many others at global scale.
My work covers a wide range of areas including high-quality multi-view stereo, surface reconstruction, structure from motion, single-view 3D modeling, 3D object detection and tracking, most of them led to publications in top-tier computer graphics and computer vision venues such as SIGGRAPH / SIGGRAPH Asia, CVPR, ECCV and ICCV. I contributed to more than 10 granted US patents and many successful tech transfers empowering commercial products, such as Douyin’s Landmark AR, Snapchat Landmarkers, Marker Tracking, 3D Stickers, which received media coverage from CNET, Engadget and TechCrunch etc.
Before joining ByteDance, I was a Lead Research Scientist at Snap Inc. working on a variety of AR research projects and products. Even before that, I worked as Research Scientist at Adobe Research on 3D scanning and 3D modeling. I obtained my PhD from Princeton University Computer Science Department in 2013 and my bachelor degree with honors from Tsinghua University School of Software in 2007.
Best way to contact me: linjie dot luo at gmail dot com
Latest News
- I am co-organizing a workshop Holistic Scene Structures for 3D Vision at ECCV 2020
- 1 paper is accepted to 3DV 2020
- 1 paper is accepted to TOG and presented at SIGGRAPH 2020
- 2 papers are accepted to ICCV 2019 (1 selected as oral presentation)
- 1 paper is accepted as oral presentation to CVPR 2019
- I will be on the International Program Committee (IPC) for Eurographics 2019
- 2 papers are conditionally accepted to SIGGRAPH Asia 2018
- 3 papers are accepted to ECCV 2018
Media Coverage
- Landmark AR (11/2019): 新华网 (Xinhuanet)
- Landmarkers (12/2018): DailyMail TheVerge Engadget Gizmodo TechCrunch
- Marker Tracking (10/2018): Engadget Adweek
- 3D Stickers (04/2016): TechCrunch Mashable BGR
- 3D Hair reconstruction (05/2013): CNET