YUXUAN ZHANG - Home

I was a CS Ph.D. student at Princeton University, where I feel fortunate to be advised by professor Felix Heide. My research interests are broadly in machine learning and computer vision.

Before joining Princeton, I worked at the NVIDIA AI lab as a deep learning researcher. During my time at NVIDIA, I feel so lucky to be mentored by professors Sanja Fidler and Antonio Torralba, who gave me so much guidance and support and showed me how to conduct great research. I also once worked as an undergraduate research assistant at the University of Waterloo, supervised by professor Florian Kerschbaum.

GOOGLE SCHOLAR | LINKEDIN | EMAIL

Publication

DatasetGAN: Efficient Labeled Data Factory with Minimal Human Effort

Yuxuan Zhang*, Huan ling*, Jun Gao, Kangxue Yin, Jean-Francois Lafleche, Adela Barriuso, Antonio Torralba, Sanja Fidler
Accepted to Conference on Computer Vision and Pattern Recognition (CVPR) 2021, Oral Presentation
arXiv: 2104.06490 | Project Page | Code

Image GANs meet Differentiable Rendering for Inverse Graphics and Interpretable 3D Neural Rendering

Yuxuan Zhang*, Wenzheng Chen*, Huan ling, Jun Gao, Yinan Zhang, Antonio Torralba, Sanja Fidler.
Accepted to International Conference on Learning Representations (ICLR) 2021, Oral Presentation
arXiv: 2010.09125 | Project Page | Released Omniverse Toolkit

All You Need is RAW: Defending Against Adversarial Attacks with Camera Image Pipelines

Yuxuan Zhang , Bo Dong, Felix Heide
Accepted to European Conference on Computer Vision (ECCV) 2022, Poster Presentation
arXiv: 2112.09219 | Project Page | Code

The Implicit Values of A Good Hand Shake: Handheld Multi-Frame Neural Depth Refinement

Ilya Chugunov, Yuxuan Zhang, Zhihao Xia, Cecilia Zhang, Jiawen Chen, Felix Heide
Accepted to Conference on Computer Vision and Pattern Recognition (CVPR) 2022, Oral Presentation
arXiv: 2111.13738 | Project Page | Code

Deep Neural Network Fingerprinting by Conferrable Adversarial Examples

Nils Lukas, Yuxuan Zhang, Florian Kerschbaum
Accepted to International Conference on Learning Representations (ICLR) 2021, Spotlight Presentation
arXiv: 1912.00888