5 papers
Circular Quasiconformal Deturbulence: Geometry-Based Restoration from Multiple Turbulent Frames
Chu Chen, Han Zhang, Lok Ming Lui
Imaging through inhomogeneous media often results in severe distortions, posing significant challenges to downstream image-processing tasks. The lack of clean paired images makes s…
Learning Task-Aware Sampling with Shared Saliency through Density-Equalizing Mappings
Tsz Lok Ip, Han Zhang, Lok Ming Lui
In image and surface-based learning tasks, convolutional features are typically extracted using receptive fields that are sampled uniformly across the entire domain. However, infor…
Structure-Preserving Neural Surrogates with Tractable Uncertainty Quantification
Handi Zhang, Adrienne M. Propp, Brooks Kinch +2
Recent advances in scientific machine learning provide a means of near-real-time solution to partial differential equations (PDEs), but lack the theoretical underpinnings of conven…
Quasi-Conformal Convolution : A Learnable Convolution for Deep Learning on Simply Connected Open Surfaces
Han Zhang, Tsz Lok Ip, Lok Ming Lui
Deep learning on non-Euclidean domains is important for analyzing complex geometric data that lacks common coordinate systems and familiar Euclidean properties. A central challenge…
Deformation-Invariant Neural Network and Its Applications in Distorted Image Restoration and Analysis
Han Zhang, Qiguang Chen, Lok Ming Lui
Images degraded by geometric distortions pose a significant challenge to imaging and computer vision tasks such as object recognition. Deep learning-based imaging models usually fa…