Biography
Jiawang Bian is a Research Scientist at ByteDance (San Jose). He focuses on advancing 3D Computer Vision research, including 3D reconstruction, generation, editing, and understanding. He received his Ph.D. from the University of Adelaide, advised by Prof. Ian Reid and Prof. Chunhua Shen, and his Bachelor’s degree from Nankai University, where he was mentored by Prof. Ming-Ming Cheng. Before joining ByteDance, Jiawang was a postdoctoral researcher at the University of Oxford, working with Prof. Philip Torr and Prof. Victor Adrian Prisacariu, and later at MBZUAI, working with Prof. Ian Reid. He also gained research experience as a research assistant at SUTD and through internships at ADSC, TuSimple, Amazon, and Meta. His long-term goal is to bridge the gap between high-quality 3D perception and real-world AI systems.
Selected Papers
- Image Matching, Depth, and Pose
- GMS: Grid-based Motion Statistics for Fast, Ultra-robust Feature correspondence, IJCV 2020 (CVPR 2017)
- An Evaluation of Feature Matchers for Fundamental Matrix Estimation, BMVC 2019
- Visual Odometry Revisited: What Should Be Learnt? ICRA 2020
- [SC-Depth] Unsupervised Scale-consistent Depth Learning from Video, IJCV 2021 (NeurIPS 2019)
- [SC-DepthV2] Auto-Rectify Network for Unsupervised Indoor Depth Estimation, TPAMI 2021
- [SC-DepthV3] Robust Self-supervised Monocular Depth Estimation for Dynamic Scenes, TPMAI 2023
- Neural Rendering and 3D Reconstruction
- NoPe-NeRF: Optimising Neural Radiance Field with No Pose Prior, CVPR 2023
- MobileBrick: Building LEGO for 3D Reconstruction on Mobile Devices, CVPR 2023
- PoRF: Pose Residual Field for Accurate Neural Surface Reconstruction, ICLR 2024
- Neural Refinement for Absolute Pose Regression with Feature Synthesis, CVPR 2024
- 3D Generation, Editing, and LLM
Featured Projects
