Gene Cheung

Image classifier learning from noisy labels via generalized graph smoothness priors

2016

When collecting samples via crowd-sourcing for semi-supervised learning, often labels that designate events of interest are assigned unreliably, resulting in label noise. In this paper, we propose a robust method for graph-based image classifier learning …

Learning Sparse Graph Laplacian with K Eigenvector Prior via Iterative Glasso and Projection

2021

Learning a suitable graph is an important precursor to many graph signal processing (GSP) pipelines, such as graph signal compression and denoising. Previous graph learning algorithms either i) make assumptions on graph connectivity (e.g., graph …

Constructing an Interpretable Deep Denoiser by Unrolling Graph Laplacian Regularizer

2024 · arXiv (Cornell University)

An image denoiser can be used for a wide range of restoration problems via the Plug-and-Play (PnP) architecture. In this paper, we propose a general framework to build an interpretable graph-based deep denoiser (GDD) by …