Yu-Jen Wei, Tsu-Tsai Wei, Tien-Ying Kuo, Po-Chyi Su
{"title":"用于图像着色的两阶段金字塔卷积神经网络","authors":"Yu-Jen Wei, Tsu-Tsai Wei, Tien-Ying Kuo, Po-Chyi Su","doi":"10.1017/ATSIP.2021.13","DOIUrl":null,"url":null,"abstract":"The development of colorization algorithms through deep learning has become the current research trend. These algorithms colorize grayscale images automatically and quickly, but the colors produced are usually subdued and have low saturation. This research addresses this issue of existing algorithms by presenting a two-stage convolutional neural network (CNN) structure with the first and second stages being a chroma map generation network and a refinement network, respectively. To begin, we convert the color space of an image from RGB to HSV to predict its low-resolution chroma components and therefore reduce the computational complexity. Following that, the first-stage output is zoomed in and its detail is enhanced with a pyramidal CNN, resulting in a colorized image. Experiments show that, while using fewer parameters, our methodology produces results with more realistic color and higher saturation than existing methods.","PeriodicalId":44812,"journal":{"name":"APSIPA Transactions on Signal and Information Processing","volume":" ","pages":""},"PeriodicalIF":3.2000,"publicationDate":"2021-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Two-stage pyramidal convolutional neural networks for image colorization\",\"authors\":\"Yu-Jen Wei, Tsu-Tsai Wei, Tien-Ying Kuo, Po-Chyi Su\",\"doi\":\"10.1017/ATSIP.2021.13\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The development of colorization algorithms through deep learning has become the current research trend. These algorithms colorize grayscale images automatically and quickly, but the colors produced are usually subdued and have low saturation. This research addresses this issue of existing algorithms by presenting a two-stage convolutional neural network (CNN) structure with the first and second stages being a chroma map generation network and a refinement network, respectively. To begin, we convert the color space of an image from RGB to HSV to predict its low-resolution chroma components and therefore reduce the computational complexity. Following that, the first-stage output is zoomed in and its detail is enhanced with a pyramidal CNN, resulting in a colorized image. Experiments show that, while using fewer parameters, our methodology produces results with more realistic color and higher saturation than existing methods.\",\"PeriodicalId\":44812,\"journal\":{\"name\":\"APSIPA Transactions on Signal and Information Processing\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":3.2000,\"publicationDate\":\"2021-10-08\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"APSIPA Transactions on Signal and Information Processing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1017/ATSIP.2021.13\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"Computer Science\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"APSIPA Transactions on Signal and Information Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1017/ATSIP.2021.13","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"Computer Science","Score":null,"Total":0}
Two-stage pyramidal convolutional neural networks for image colorization
The development of colorization algorithms through deep learning has become the current research trend. These algorithms colorize grayscale images automatically and quickly, but the colors produced are usually subdued and have low saturation. This research addresses this issue of existing algorithms by presenting a two-stage convolutional neural network (CNN) structure with the first and second stages being a chroma map generation network and a refinement network, respectively. To begin, we convert the color space of an image from RGB to HSV to predict its low-resolution chroma components and therefore reduce the computational complexity. Following that, the first-stage output is zoomed in and its detail is enhanced with a pyramidal CNN, resulting in a colorized image. Experiments show that, while using fewer parameters, our methodology produces results with more realistic color and higher saturation than existing methods.