Rethinking RGB Color Representation for Image Restoration Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lee, Jaerin, Park, JoonKyu, Baik, Sungyong, Lee, Kyoung Mu
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910318983118848
author Lee, Jaerin
Park, JoonKyu
Baik, Sungyong
Lee, Kyoung Mu
author_facet Lee, Jaerin
Park, JoonKyu
Baik, Sungyong
Lee, Kyoung Mu
contents Image restoration models are typically trained with a pixel-wise distance loss defined over the RGB color representation space, which is well known to be a source of blurry and unrealistic textures in the restored images. The reason, we believe, is that the three-channel RGB space is insufficient for supervising the restoration models. To this end, we augment the representation to hold structural information of local neighborhoods at each pixel while keeping the color information and pixel-grainedness unharmed. The result is a new representation space, dubbed augmented RGB ($a$RGB) space. Substituting the underlying representation space for the per-pixel losses facilitates the training of image restoration models, thereby improving the performance without affecting the evaluation phase. Notably, when combined with auxiliary objectives such as adversarial or perceptual losses, our $a$RGB space consistently improves overall metrics by reconstructing both color and local structures, overcoming the conventional perception-distortion trade-off.
format Preprint
id arxiv_https___arxiv_org_abs_2402_03399
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Rethinking RGB Color Representation for Image Restoration Models
Lee, Jaerin
Park, JoonKyu
Baik, Sungyong
Lee, Kyoung Mu
Image and Video Processing
Computer Vision and Pattern Recognition
Image restoration models are typically trained with a pixel-wise distance loss defined over the RGB color representation space, which is well known to be a source of blurry and unrealistic textures in the restored images. The reason, we believe, is that the three-channel RGB space is insufficient for supervising the restoration models. To this end, we augment the representation to hold structural information of local neighborhoods at each pixel while keeping the color information and pixel-grainedness unharmed. The result is a new representation space, dubbed augmented RGB ($a$RGB) space. Substituting the underlying representation space for the per-pixel losses facilitates the training of image restoration models, thereby improving the performance without affecting the evaluation phase. Notably, when combined with auxiliary objectives such as adversarial or perceptual losses, our $a$RGB space consistently improves overall metrics by reconstructing both color and local structures, overcoming the conventional perception-distortion trade-off.
title Rethinking RGB Color Representation for Image Restoration Models
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2402.03399