I'm a second-year CSE Ph.D. student at the University of Michigan, advised by David Fouhey. I obtained my bachelor degree from both the University of Michigan and Shanghai Jiao Tong University. During my undergraduate study at Michigan, I worked with Jia Deng in Vision & Learning Lab.
My primary research interest lies in Computer Vision, espeically 3D vision and recognition. I've closely collaborated with Weifeng Chen. I'm proud to work with my labmates in Fouhey AI Lab (FAIL) and the computer vision group @ Michigan.
- [2020/08] Associative3D is invited to present at ECCV 2020 Workshop Holistic Scene Structures for 3D Vision.
- [2020/07] "Associative3D: Volumetric Reconstruction from Sparse Views" is accepted at ECCV 2020!
- [2020/02] "OASIS: A Large-Scale Dataset for Single-Image 3D in the Wild" is accepted at CVPR 2020!
My research focuses on 3D vision, i.e. how do we recover the 3D world from a 2D image? I'm particularly interested in recovering 3D representations from Internet videos and computer graphics. Images above demonstrate different 3D representations we extract from 2D image(s) in my research projects.
Instructional Aide / Teaching Assistant:
- EECS 442 Computer Vision, Winter 2019, University of Michigan. With David Fouhey.
- VE280 Programming & Data Structures, Summer 2019, Shanghai Jiao Tong University. With Weikang Qian and Paul Weng.
We present Associative3D, which addresses 3D volumetric reconstruction from two views of a scene with an unknown camera, by simultaneously reconstructing objects and figuring out their relationship.
We present Open Annotations of Single Image Surfaces (OASIS), a dataset for single-image 3D in the wild consisting of dense annotations of detailed 3D geometry for Internet images.
We propose a method to automatically generate training data for single-view depth through Structure-from-Motion (SfM) on Internet videos.