Tianwen Fu
I am a second-year PhD student in computer science at the University of Southern California. Previously, I graduated from the master’s program in computer vision at Carnegie Mellon University, where I was honored to work on a computational photography project with event cameras with Professor Ioannis Gkioulekas. I did my capstone project at the R-PAD lab led by Professor David Held on 3D scene understanding.
I obtained a Bachelor of Science in Mathematics and Information Engineering with first-class honors from the Chinese University of Hong Kong, where I had the pleasure to work on computer graphics with Professor Chi-Wing Fu. I was also honored to work as a research intern at SenseTime with Professor Jifeng Dai and Dr. Hao Li.
Publications
- RDD: Robust Feature Detector and Descriptor using Deformable TransformerIn Proceedings of the Computer Vision and Pattern Recognition Conference, 2025
- Unraveling the Complexities of Simplicity Bias: Mitigating and Amplifying FactorsIn NeurIPS 2023 Workshop on Mathematics of Modern Machine Learning
- Autoloss-zero: Searching loss functions from scratch for generic tasksIn Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022
- Computational design and optimization of non-circular gearsIn Computer Graphics Forum, 2020