NeurTV: Total Variation on the Neural Domain

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Luo, Yisi, Zhao, Xile, Ye, Kai, Meng, Deyu
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909445093588992
author Luo, Yisi
Zhao, Xile
Ye, Kai
Meng, Deyu
author_facet Luo, Yisi
Zhao, Xile
Ye, Kai
Meng, Deyu
contents Recently, we have witnessed the success of total variation (TV) for many imaging applications. However, traditional TV is defined on the original pixel domain, which limits its potential. In this work, we suggest a new TV regularization defined on the neural domain. Concretely, the discrete data is implicitly and continuously represented by a deep neural network (DNN), and we use the derivatives of DNN outputs w.r.t. input coordinates to capture local correlations of data. As compared with classical TV on the original domain, the proposed TV on the neural domain (termed NeurTV) enjoys the following advantages. First, NeurTV is free of discretization error induced by the discrete difference operator. Second, NeurTV is not limited to meshgrid but is suitable for both meshgrid and non-meshgrid data. Third, NeurTV can more exactly capture local correlations across data for any direction and any order of derivatives attributed to the implicit and continuous nature of neural domain. We theoretically reinterpret NeurTV under the variational approximation framework, which allows us to build the connection between NeurTV and classical TV and inspires us to develop variants (e.g., space-variant NeurTV). Extensive numerical experiments with meshgrid data (e.g., color and hyperspectral images) and non-meshgrid data (e.g., point clouds and spatial transcriptomics) showcase the effectiveness of the proposed methods.
format Preprint
id arxiv_https___arxiv_org_abs_2405_17241
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle NeurTV: Total Variation on the Neural Domain
Luo, Yisi
Zhao, Xile
Ye, Kai
Meng, Deyu
Computer Vision and Pattern Recognition
Image and Video Processing
94A08, 68U10, 68T45
Recently, we have witnessed the success of total variation (TV) for many imaging applications. However, traditional TV is defined on the original pixel domain, which limits its potential. In this work, we suggest a new TV regularization defined on the neural domain. Concretely, the discrete data is implicitly and continuously represented by a deep neural network (DNN), and we use the derivatives of DNN outputs w.r.t. input coordinates to capture local correlations of data. As compared with classical TV on the original domain, the proposed TV on the neural domain (termed NeurTV) enjoys the following advantages. First, NeurTV is free of discretization error induced by the discrete difference operator. Second, NeurTV is not limited to meshgrid but is suitable for both meshgrid and non-meshgrid data. Third, NeurTV can more exactly capture local correlations across data for any direction and any order of derivatives attributed to the implicit and continuous nature of neural domain. We theoretically reinterpret NeurTV under the variational approximation framework, which allows us to build the connection between NeurTV and classical TV and inspires us to develop variants (e.g., space-variant NeurTV). Extensive numerical experiments with meshgrid data (e.g., color and hyperspectral images) and non-meshgrid data (e.g., point clouds and spatial transcriptomics) showcase the effectiveness of the proposed methods.
title NeurTV: Total Variation on the Neural Domain
topic Computer Vision and Pattern Recognition
Image and Video Processing
94A08, 68U10, 68T45
url https://arxiv.org/abs/2405.17241