CV
Education
- Aug. 2010 - Jun. 2016, Ph.D., Tsinghua University, Department of Computer Science & Technology
- Excellent Graduate of CS Department: I published 17 peer-reviewed papers in the field of computer networking during my Ph.D. program, and 9 of them are first author papers. I also contribute to important scientific projects, including 863 projects, etc.
- Thesis: Research on the Key Issues of Usability of Named Data Networking, supervised by Prof. Jun Bi
- Sep. 2006 - Jun. 2010, B.S.,Tsinghua University, School of Software.
- Thesis: Research on the Evolution of IPv6, supervised by Prof. Jun Bi and Prof. Fei He
- Jul. 2014 - Jul. 2015, Joint PhD Program, University of California Los Angeles(UCLA), supervised by Prof. Lixia Zhang
- ICCCN’17 paper: NDNS: A DNS-Like Name Service for NDN
- Open Source Project: ndns, which is the DNS system for NDN testbed
- Aug. 2011 - Jul 2012, Tsinghua-Berkeley Global Technology Entrepreneurship Program
- served as tech leader of 2nd Award Team
Work experience
- 2022 - now, IDEA, Senior CV Researcher
- Open Set Detection, esp. prompt tuning to customize detection targets given labelled dataset and pretrained model, keypoint detection, etc.
- keypoints detection section in Report, DINO-X: A Unified Vision Model for Open-World Object Detection and Understanding
- supervised prompt tuning part in Blog, Grounding DINO 1.6
- edge model part in Report, Grounding DINO 1.5: Advance the “Edge” of Open-Set Object Detection
- Keypoint Detection, esp. keypoints detection in the wild
- CVPR’25 paper: HandOS: 3D Hand Reconstruction in One Stage
- 3D reconstruction and Novel View Synthesis:
- CVPR’25 paper: LeanGaussian: Breaking Pixel or Point Cloud Correspondence in Modeling 3D Gaussians
- paper: UniG: Modelling Unitary 3D Gaussians for View-consistent 3D Reconstruction
- ongoing work: JocSplat: Joint Optimization for Camera parameters and 3D Gaussians, which allows 3D reconstuction without the need for camera extrinsics and intrinsics
- Inertial Navigation and Integrated Localization (IMU + GPS + wheel odom)
- Open Set Detection, esp. prompt tuning to customize detection targets given labelled dataset and pretrained model, keypoint detection, etc.
- 2018 - 2022, SenseTime, Senior Researcher. I was awarded the “Excellent Employee of SenseTime” in 2019. To cope with challenges in the AI-based manifuactor, I worked on the following topics:
- Defect Detection
- Structured Light Imaging
- Optics and Lasers in Engineering (2023) paper, Defect classification for specular surfaces based on deflectometry and multi-modal fusion network
- Defect Detection in Automobile Smart Manufacturing
- Structured Light Imaging
- 6D pose estimation for robotic manipulation
- TMM’25 paper, Geo6D: Geometric-Constraints-Guided Direct Object 6D Pose Estimation Network
- CVPR’22 oral paper, Uni6D: A Unified CNN Framework without Projection Breakdown for 6D Pose Estimation
- AISTATS’23 paper, Uni6Dv2: Noise Elimination for 6D Pose Estimation
- Person Reidentification
- AI automiation in Transportation System
- Defect Detection
- 2016 - 2018, Kandao technology, Senior Engineer. I was awarded the “Excellent Employee of Kandao” in 2017, and worked on the following topics:
- VR Video Streaming Optimization and Video Content Delivery and Distribution (CDN) System
- SIGCOMM’18 VR Workshop paper: Smart Streaming of Panoramic Video
- patents (China and US): Panoramic video picture quality display method and device
- patents (China and US): Method and device for picture quality assessment after dynamic panoramic video stream cropping
- VR Video Streaming Optimization and Video Content Delivery and Distribution (CDN) System
Skills
- Mathematics
- Linear Algebra
- Lie Algebra/Rotation Group
- Probability Theory
- Taylor Expansion
- Optimization Theory (Basic)
- Programming:
- Languages: Python, GoLang, C++
- tools: Linux, git-based work flow, make, gcc, ffmpeg
- libraries: pytorch, mmengine, pytorch-lightning, opencv, scikit-learn, matplotlib
Hobbies:
- Badminton
- Hiking
- Programming to solve real-world problems
- Reading Novels for relaxation and imagination
- Philosophy to explore the nature of the world and meaning of life