Search Results for author: Gene Cheung

Found 44 papers, 5 papers with code

Interpretable Lightweight Transformer via Unrolling of Learned Graph Smoothness Priors

no code implementations6 Jun 2024 Tam Thuc Do, Parham Eftekhar, Seyed Alireza Hosseini, Gene Cheung, Philip Chou

We build interpretable and lightweight transformer-like neural networks by unrolling iterative optimization algorithms that minimize graph smoothness priors -- the quadratic graph Laplacian regularizer (GLR) and the $\ell_1$-norm graph total variation (GTV) -- subject to an interpolation constraint.

Graph Learning

Signal Processing in the Retina: Interpretable Graph Classifier to Predict Ganglion Cell Responses

no code implementations3 Jan 2024 Yasaman Parhizkar, Gene Cheung, Andrew W. Eckford

To extract knowledge from the cell firings, in this paper we learn an interpretable graph-based classifier from data to predict the firings of ganglion cells in response to visual stimuli.

Binary Classification Metric Learning

Learned Nonlinear Predictor for Critically Sampled 3D Point Cloud Attribute Compression

no code implementations22 Nov 2023 Tam Thuc Do, Philip A. Chou, Gene Cheung

We study 3D point cloud attribute compression via a volumetric approach: assuming point cloud geometry is known at both encoder and decoder, parameters $\theta$ of a continuous attribute function $f: \mathbb{R}^3 \mapsto \mathbb{R}$ are quantized to $\hat{\theta}$ and encoded, so that discrete samples $f_{\hat{\theta}}(\mathbf{x}_i)$ can be recovered at known 3D points $\mathbf{x}_i \in \mathbb{R}^3$ at the decoder.

Attribute Decoder

Volumetric 3D Point Cloud Attribute Compression: Learned polynomial bilateral filter for prediction

no code implementations22 Nov 2023 Tam Thuc Do, Philip A. Chou, Gene Cheung

We extend a previous study on 3D point cloud attribute compression scheme that uses a volumetric approach: given a target volumetric attribute function $f : \mathbb{R}^3 \mapsto \mathbb{R}$, we quantize and encode parameters $\theta$ that characterize $f$ at the encoder, for reconstruction $f_{\hat{\theta}}(\mathbf(x))$ at known 3D points $\mathbf(x)$ at the decoder.

Attribute Decoder

Complex Graph Laplacian Regularizer for Inferencing Grid States

no code implementations4 Jul 2023 Chinthaka Dinesh, Junfei Wang, Gene Cheung, Pirathayini Srikantha

In order to maintain stable grid operations, system monitoring and control processes require the computation of grid states (e. g. voltage magnitude and angles) at high granularity.

Graph Sparsification for GCN Towards Optimal Crop Yield Predictions

no code implementations2 Jun 2023 Saghar Bagheri, Gene Cheung, Tim Eadie

Specifically, we first show that greedily removing an edge at a time that induces the minimal change in the second eigenvalue leads to a sparse graph with good GCN performance.

Crop Yield Prediction

Volumetric Attribute Compression for 3D Point Clouds using Feedforward Network with Geometric Attention

no code implementations1 Apr 2023 Tam Thuc Do, Philip A. Chou, Gene Cheung

We study 3D point cloud attribute compression using a volumetric approach: given a target volumetric attribute function $f : \mathbb{R}^3 \rightarrow \mathbb{R}$, we quantize and encode parameter vector $\theta$ that characterizes $f$ at the encoder, for reconstruction $f_{\hat{\theta}}(\mathbf{x})$ at known 3D points $\mathbf{x}$'s at the decoder.

Attribute Decoder

Efficient Directed Graph Sampling via Gershgorin Disc Alignment

no code implementations25 Oct 2022 Yuejiang Li, Hong Vicky Zhao, Gene Cheung

To minimize worst-case reconstruction error of the linear system solution $\mathbf{x}^* = \mathbf{C}^{-1} \mathbf{H}^\top \mathbf{y}$ with symmetric coefficient matrix $\mathbf{C} = \mathbf{H}^\top \mathbf{H} + \mu \mathbf{L}_{rw}^\top \mathbf{L}_{rw}$, the sampling objective is to choose $\mathbf{H}$ to maximize the smallest eigenvalue $\lambda_{\min}(\mathbf{C})$ of $\mathbf{C}$.

Graph Sampling

Efficient Signed Graph Sampling via Balancing & Gershgorin Disc Perfect Alignment

no code implementations18 Aug 2022 Chinthaka Dinesh, Gene Cheung, Saghar Bagheri, Ivan V. Bajic

Experimental results show that our signed graph sampling method outperformed existing fast sampling schemes noticeably on various datasets.

Graph Sampling

Unsupervised Graph Spectral Feature Denoising for Crop Yield Prediction

no code implementations4 Aug 2022 Saghar Bagheri, Chinthaka Dinesh, Gene Cheung, Timothy Eadie

Prediction of annual crop yields at a county granularity is important for national food production and price stability.

Crop Yield Prediction Denoising +1

Manifold Graph Signal Restoration using Gradient Graph Laplacian Regularizer

no code implementations9 Jun 2022 Fei Chen, Gene Cheung, Xue Zhang

In this paper, focusing on manifold graphs -- collections of uniform discrete samples on low-dimensional continuous manifolds -- we generalize GLR to gradient graph Laplacian regularizer (GGLR) that promotes planar / piecewise planar (PWP) signal reconstruction.

Graph Embedding

Hybrid Model-based / Data-driven Graph Transform for Image Coding

no code implementations2 Mar 2022 Saghar Bagheri, Tam Thuc Do, Gene Cheung, Antonio Ortega

Transform coding to sparsify signal representations remains crucial in an image compression pipeline.

Graph Learning Image Compression

Sparse Graph Learning with Spectrum Prior for Deep Graph Convolutional Networks

no code implementations28 Feb 2022 Jin Zeng, Yang Liu, Gene Cheung, Wei Hu

Specifically, based on a spectral analysis of multilayer GCN output, we derive a spectrum prior for the graph Laplacian matrix $\mathbf{L}$ to robustify the model expressiveness against over-smoothing.

Graph Learning

Fast Computation of Generalized Eigenvectors for Manifold Graph Embedding

no code implementations15 Dec 2021 Fei Chen, Gene Cheung, Xue Zhang

Experiments show that our embedding is among the fastest in the literature, while producing the best clustering performance for manifold graphs.

Clustering Graph Embedding

Graph-Based Depth Denoising & Dequantization for Point Cloud Enhancement

no code implementations9 Nov 2021 Xue Zhang, Gene Cheung, Jiahao Pang, Yash Sanghvi, Abhiram Gnanasambandam, Stanley H. Chan

Specifically, we model depth formation as a combined process of signal-dependent noise addition and non-uniform log-based quantization.

Graph Learning Image Denoising +1

Fast Graph Sampling for Short Video Summarization using Gershgorin Disc Alignment

no code implementations21 Oct 2021 Sadid Sahami, Gene Cheung, Chia-Wen Lin

We prove that, after partitioning $\mathcal{G}$ into $Q$ sub-graphs $\{\mathcal{G}^q\}^Q_{q=1}$, the smallest Gershgorin circle theorem (GCT) lower bound of $Q$ corresponding coefficient matrices -- $\min_q \lambda^-_{\min}(\mathbf{B}^q)$ -- is a lower bound for $\lambda_{\min}(\mathbf{B})$.

Graph Sampling Video Summarization

Fast sensor placement by enlarging principle submatrix for large-scale linear inverse problems

no code implementations6 Oct 2021 Fen Wang, Gene Cheung, Taihao Li, Ying Du, Yu-Ping Ruan

Sensor placement for linear inverse problems is the selection of locations to assign sensors so that the entire physical signal can be well recovered from partial observations.

Unfolding Projection-free SDP Relaxation of Binary Graph Classifier via GDPA Linearization

no code implementations10 Sep 2021 Cheng Yang, Gene Cheung, Wai-tian Tan, Guangtao Zhai

Algorithm unfolding creates an interpretable and parsimonious neural network architecture by implementing each iteration of a model-based algorithm as a neural layer.

Projection-free Graph-based Classifier Learning using Gershgorin Disc Perfect Alignment

no code implementations NeurIPS 2021 Cheng Yang, Gene Cheung, Guangtao Zhai

We repose the SDR dual for solution $\bar{\mathbf{H}}$, then replace the PSD cone constraint $\bar{\mathbf{H}} \succeq 0$ with linear constraints derived from GDPA -- sufficient conditions to ensure $\bar{\mathbf{H}}$ is PSD -- so that the optimization becomes an LP per iteration.

Point Cloud Sampling via Graph Balancing and Gershgorin Disc Alignment

no code implementations10 Mar 2021 Chinthaka Dinesh, Gene Cheung, Ivan Bajic

Specifically, to articulate a sampling objective, we first assume a super-resolution (SR) method based on feature graph Laplacian regularization (FGLR) that reconstructs the original high-resolution PC, given 3D points chosen by a sampling matrix $\H$.

Graph Sampling Object Recognition +1

Pre-demosaic Graph-based Light Field Image Compression

no code implementations15 Feb 2021 Yung-Hsuan Chao, Haoran Hong, Gene Cheung, Antonio Ortega

Using a conventional Bayer pattern, data captured at each pixel is a single color component (R, G or B). The sensed data then undergoes demosaicking (interpolation of RGB components per pixel) and conversion to an array of sub-aperture images (SAIs).

Demosaicking Graph Learning +1

Fast & Robust Image Interpolation using Gradient Graph Laplacian Regularizer

no code implementations25 Jan 2021 Fei Chen, Gene Cheung, Xue Zhang

In the graph signal processing (GSP) literature, it has been shown that signal-dependent graph Laplacian regularizer (GLR) can efficiently promote piecewise constant (PWC) signal reconstruction for various image restoration tasks.

Image Restoration

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

no code implementations25 Oct 2020 Saghar Bagheri, Gene Cheung, Antonio Ortega, Fen Wang

Learning a suitable graph is an important precursor to many graph signal processing (GSP) pipelines, such as graph spectral signal compression and denoising.

Denoising Graph Learning

Unrolling of Deep Graph Total Variation for Image Denoising

1 code implementation21 Oct 2020 Huy Vu, Gene Cheung, Yonina C. Eldar

While deep learning (DL) architectures like convolutional neural networks (CNNs) have enabled effective solutions in image denoising, in general their implementations overly rely on training data, lack interpretability, and require tuning of a large parameter set.

Image Denoising Rolling Shutter Correction

Signed Graph Metric Learning via Gershgorin Disc Perfect Alignment

1 code implementation15 Jun 2020 Cheng Yang, Gene Cheung, Wei Hu

Given a convex and differentiable objective $Q(\M)$ for a real symmetric matrix $\M$ in the positive definite (PD) cone -- used to compute Mahalanobis distances -- we propose a fast general metric learning framework that is entirely projection-free.

Binary Classification Metric Learning

Sampling Signals on Graphs: From Theory to Applications

no code implementations9 Mar 2020 Yuichi Tanaka, Yonina C. Eldar, Antonio Ortega, Gene Cheung

In this article, we review current progress on sampling over graphs focusing on theory and potential applications.

Graph Metric Learning via Gershgorin Disc Alignment

no code implementations28 Jan 2020 Cheng Yang, Gene Cheung, Wei Hu

We propose a fast general projection-free metric learning framework, where the minimization objective $\min_{\textbf{M} \in \mathcal{S}} Q(\textbf{M})$ is a convex differentiable function of the metric matrix $\textbf{M}$, and $\textbf{M}$ resides in the set $\mathcal{S}$ of generalized graph Laplacian matrices for connected graphs with positive edge weights and node degrees.

Metric Learning

Robust Deep Graph Based Learning for Binary Classification

no code implementations6 Dec 2019 Minxiang Ye, Vladimir Stankovic, Lina Stankovic, Gene Cheung

In this paper, we propose a robust binary classifier, based on CNNs, to learn deep metric functions, which are then used to construct an optimal underlying graph structure used to clean noisy labels via graph Laplacian regularization (GLR).

Binary Classification Classification +1

Feature Graph Learning for 3D Point Cloud Denoising

no code implementations22 Jul 2019 Wei Hu, Xiang Gao, Gene Cheung, Zongming Guo

In this work, we assume instead the availability of a relevant feature vector $\mathbf{f}_i$ per node $i$, from which we compute an optimal feature graph via optimization of a feature metric.

Graph Learning Image Denoising

Deep Graph Laplacian Regularization for Robust Denoising of Real Images

1 code implementation31 Jul 2018 Jin Zeng, Jiahao Pang, Wenxiu Sun, Gene Cheung

In this work, we combine the robustness merit of model-based approaches and the learning power of data-driven approaches for real image denoising.

Domain Generalization Image Denoising +1

SiGAN: Siamese Generative Adversarial Network for Identity-Preserving Face Hallucination

1 code implementation22 Jul 2018 Chih-Chung Hsu, Chia-Wen Lin, Weng-Tai Su, Gene Cheung

Despite generative adversarial networks (GANs) can hallucinate photo-realistic high-resolution (HR) faces from low-resolution (LR) faces, they cannot guarantee preserving the identities of hallucinated HR faces, making the HR faces poorly recognizable.

Face Hallucination Face Reconstruction +3

3D Point Cloud Denoising using Graph Laplacian Regularization of a Low Dimensional Manifold Model

no code implementations20 Mar 2018 Jin Zeng, Gene Cheung, Michael Ng, Jiahao Pang, Cheng Yang

Due to discrete observations of the patches on the manifold, we approximate the manifold dimension computation defined in the continuous domain with a patch-based graph Laplacian regularizer and propose a new discrete patch distance measure to quantify the similarity between two same-sized surface patches for graph construction that is robust to noise.

Denoising graph construction +2

Graph-Based Blind Image Deblurring From a Single Photograph

no code implementations22 Feb 2018 Yuanchao Bai, Gene Cheung, Xian-Ming Liu, Wen Gao

We leverage the new graph spectral interpretation for RGTV to design an efficient algorithm that solves for the skeleton image and the blur kernel alternately.

Blind Image Deblurring Image Deblurring

Non-Local Graph-Based Prediction For Reversible Data Hiding In Images

no code implementations20 Feb 2018 Qi Chang, Gene Cheung, Yao Zhao, Xiaolong Li, Rongrong Ni

If sufficiently smooth, we pose a maximum a posteriori (MAP) problem using either a quadratic Laplacian regularizer or a graph total variation (GTV) term as signal prior.

Blind Image Deblurring via Reweighted Graph Total Variation

no code implementations24 Dec 2017 Yuanchao Bai, Gene Cheung, Xian-Ming Liu, Wen Gao

The problem can be solved in two parts: i) estimate a blur kernel from the blurry image, and ii) given estimated blur kernel, de-convolve blurry input to restore the target image.

Blind Image Deblurring Image Deblurring

Joint Denoising / Compression of Image Contours via Shape Prior and Context Tree

no code implementations30 Apr 2017 Amin Zheng, Gene Cheung, Dinei Florencio

We first prove theoretically that in general a joint denoising / compression approach can outperform a separate two-stage approach that first denoises then encodes contours lossily.

Action Recognition Denoising +3

Graph Fourier Transform with Negative Edges for Depth Image Coding

no code implementations10 Feb 2017 Weng-Tai Su, Gene Cheung, Chia-Wen Lin

Recent advent in graph signal processing (GSP) has led to the development of new graph-based transforms and wavelets for image / video coding, where the underlying graph describes inter-pixel correlations.

Robust Semi-Supervised Graph Classifier Learning with Negative Edge Weights

no code implementations15 Nov 2016 Gene Cheung, Weng-Tai Su, Yu Mao, Chia-Wen Lin

In response, we derive an optimal perturbation matrix $\boldsymbol{\Delta}$ - based on a fast lower-bound computation of the minimum eigenvalue of $\mathbf{L}$ via a novel application of the Haynsworth inertia additivity formula---so that $\mathbf{L} + \boldsymbol{\Delta}$ is positive semi-definite, resulting in a stable signal prior.

Random Walk Graph Laplacian based Smoothness Prior for Soft Decoding of JPEG Images

no code implementations7 Jul 2016 Xianming Liu, Gene Cheung, Xiaolin Wu, Debin Zhao

In this paper, we combine three image priors---Laplacian prior for DCT coefficients, sparsity prior and graph-signal smoothness prior for image patches---to construct an efficient JPEG soft decoding algorithm.

Clustering Image Reconstruction +1

Graph Laplacian Regularization for Image Denoising: Analysis in the Continuous Domain

no code implementations27 Apr 2016 Jiahao Pang, Gene Cheung

Focusing on image denoising, we derive an optimal metric space assuming non-local self-similarity of pixel patches, leading to an optimal graph Laplacian regularizer for denoising in the discrete domain.

Image Denoising

Precision Enhancement of 3D Surfaces from Multiple Compressed Depth Maps

no code implementations25 Feb 2014 Pengfei Wan, Gene Cheung, Philip A. Chou, Dinei Florencio, Cha Zhang, Oscar C. Au

In texture-plus-depth representation of a 3D scene, depth maps from different camera viewpoints are typically lossily compressed via the classical transform coding / coefficient quantization paradigm.

Quantization

Navigation domain representation for interactive multiview imaging

no code implementations18 Oct 2012 Thomas Maugey, Ismael Daribo, Gene Cheung, Pascal Frossard

In this paper, we propose a novel multiview data representation that permits to satisfy bandwidth and storage constraints in an interactive multiview streaming system.

Decoder Navigate

Cannot find the paper you are looking for? You can Submit a new open access paper.