Kaifeng Zhang

I am a first-year Ph.D. student in computer science at UIUC, advised by Prof. Yunzhu Li. Previously, I was an undergraduate student at Tsinghua University majoring in computer science (Yao class).

My research interest is in computer vision, machine learning and robotics. During my undergrad, I am very fortunate to have worked with Prof. Xiaolong Wang from UC San Diego, and Prof. Yang Gao from Tsinghua University.

Email / Google Scholar / Github / Twitter

News

[2023.6] Graduated from Tsinghua University. Starting my Ph.D. study at UIUC.

[2023.1] One paper gets accepted to ICLR 2023.

[2022.7] One paper gets accepted to ECCV 2022 as oral presentation.

Publications

Self-Supervised Geometric Correspondence for Category-Level 6D Object Pose Estimation in the Wild
Kaifeng Zhang, Yang Fu, Shubhankar Borse, Hong Cai, Fatih Porikli, Xiaolong Wang
International Conference on Learning Representations (ICLR), 2023.
project page / arXiv / pdf / code

We propose a self-supervised method for category-level 6D object pose estimation by learning dense 2D-3D geometric correspondences. Our method can train on in-the-wild RGB/RGBD image collections without any 3D annotations.

Semantic-Aware Fine-Grained Correspondence
Yingdong Hu, Renhao Wang, Kaifeng Zhang, Yang Gao
European Conference on Computer Vision (ECCV), 2022 (Oral).
arXiv / pdf / code

We show that fine-grained features learned with low-level contrastive objectives are complementary to semantic features from image-level SSL methods. Fusing these features can significantly improve the performance for downstream tasks.

Borrowed from Jon Barron