Di Qiu

I am a fourth year PhD student at The Chinese University of Hong Kong, where I am working on geometric modelling, computer vision and computational photography. This fall I will be working as a (virtual) intern at Google Research.

Prviously I worked as an research intern at SenseTime Research, working on applications of deep learning in computational photography with depth sensing devices and physically accurate synthetic data.

Blog  /  Email  /  Google Scholar  /  LinkedIn


I'm interested in geometric modelling in computer graphics, computer vision, machine learning, and the mathematics related to them.

Modal Uncertainty Estimation via Discrete Latent Representation

Di Qiu, Lok Ming Lui
Preprint, 2020

Learning latent mode hypothesis and their uncertainty estimation for one-to-many mappings.

Towards Geometry Guided Neural Relighting with Flash Photography

Di Qiu, Jin Zeng, Zhanghan Ke, Wenxiu Sun, Chengxi Yang
Preprint, 2020

Directional relighting from a single co-located flash image and its depth map.

Inconsistent Surface Registration via Optimization of Mapping Distortions

Di Qiu, Lok Ming Lui
Journal of Scientific Computing, 2020

Simutaneouly finding domain of correspondence and registration by optimizing distortions in the mapping differential.

Deep End-to-End Alignment and Refinement for Time-of-Flight RGB-D Modules

Di Qiu, Jiahao Pang, Chengxi Yang, Wenxiu Sun
ICCV, 2019

Cross-modal flow estimation and Time-of-Flight depth refinement using deep learning.

[code & dataset]
Computing Quasiconformal Folds

Di Qiu, Ka Chun Lam, Lok Ming Lui
SIAM Journal of Imaging Science, 2019

Computing folding and unfolding maps via a generalized form of quasiconformal mapping and crease geometry inference.


This theme is adapted from Jon Barron's page .