基于内禀分解的深度混合真实与合成训练

Sai Bi, N. Kalantari, R. Ramamoorthi
{"title":"基于内禀分解的深度混合真实与合成训练","authors":"Sai Bi, N. Kalantari, R. Ramamoorthi","doi":"10.2312/SRE.20181172","DOIUrl":null,"url":null,"abstract":"Intrinsic image decomposition is the process of separating the reflectance and shading layers of an image, which is a challenging and underdetermined problem. In this paper, we propose to systematically address this problem using a deep convolutional neural network (CNN). Although deep learning (DL) has been recently used to handle this application, the current DL methods train the network only on synthetic images as obtaining ground truth reflectance and shading for real images is difficult. Therefore, these methods fail to produce reasonable results on real images and often perform worse than the non-DL techniques. We overcome this limitation by proposing a novel hybrid approach to train our network on both synthetic and real images. Specifically, in addition to directly supervising the network using synthetic images, we train the network by enforcing it to produce the same reflectance for a pair of images of the same real-world scene with different illuminations. Furthermore, we improve the results by incorporating a bilateral solver layer into our system during both training and test stages. Experimental results show that our approach produces better results than the state-of-the-art DL and non-DL methods on various synthetic and real datasets both visually and numerically.","PeriodicalId":363391,"journal":{"name":"Eurographics Symposium on Rendering","volume":"20 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"26","resultStr":"{\"title\":\"Deep Hybrid Real and Synthetic Training for Intrinsic Decomposition\",\"authors\":\"Sai Bi, N. Kalantari, R. Ramamoorthi\",\"doi\":\"10.2312/SRE.20181172\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Intrinsic image decomposition is the process of separating the reflectance and shading layers of an image, which is a challenging and underdetermined problem. In this paper, we propose to systematically address this problem using a deep convolutional neural network (CNN). Although deep learning (DL) has been recently used to handle this application, the current DL methods train the network only on synthetic images as obtaining ground truth reflectance and shading for real images is difficult. Therefore, these methods fail to produce reasonable results on real images and often perform worse than the non-DL techniques. We overcome this limitation by proposing a novel hybrid approach to train our network on both synthetic and real images. Specifically, in addition to directly supervising the network using synthetic images, we train the network by enforcing it to produce the same reflectance for a pair of images of the same real-world scene with different illuminations. Furthermore, we improve the results by incorporating a bilateral solver layer into our system during both training and test stages. Experimental results show that our approach produces better results than the state-of-the-art DL and non-DL methods on various synthetic and real datasets both visually and numerically.\",\"PeriodicalId\":363391,\"journal\":{\"name\":\"Eurographics Symposium on Rendering\",\"volume\":\"20 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2018-07-30\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"26\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Eurographics Symposium on Rendering\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.2312/SRE.20181172\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Eurographics Symposium on Rendering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.2312/SRE.20181172","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 26

摘要

图像的内在分解是将图像的反射层和阴影层分离的过程,是一个具有挑战性和不确定性的问题。在本文中,我们建议使用深度卷积神经网络(CNN)系统地解决这个问题。虽然深度学习(DL)最近被用于处理这一应用,但目前的DL方法仅在合成图像上训练网络,因为很难获得真实图像的真实反射率和阴影。因此,这些方法不能在真实图像上产生合理的结果,并且通常比非深度学习技术表现得更差。我们克服了这一限制,提出了一种新的混合方法,在合成图像和真实图像上训练我们的网络。具体来说,除了使用合成图像直接监督网络外,我们还通过强制其对具有不同照明的相同现实场景的一对图像产生相同的反射率来训练网络。此外,我们通过在训练和测试阶段将双边求解器层合并到我们的系统中来改进结果。实验结果表明,在各种合成数据集和真实数据集上,我们的方法在视觉和数值上都比最先进的深度学习和非深度学习方法产生更好的结果。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Deep Hybrid Real and Synthetic Training for Intrinsic Decomposition
Intrinsic image decomposition is the process of separating the reflectance and shading layers of an image, which is a challenging and underdetermined problem. In this paper, we propose to systematically address this problem using a deep convolutional neural network (CNN). Although deep learning (DL) has been recently used to handle this application, the current DL methods train the network only on synthetic images as obtaining ground truth reflectance and shading for real images is difficult. Therefore, these methods fail to produce reasonable results on real images and often perform worse than the non-DL techniques. We overcome this limitation by proposing a novel hybrid approach to train our network on both synthetic and real images. Specifically, in addition to directly supervising the network using synthetic images, we train the network by enforcing it to produce the same reflectance for a pair of images of the same real-world scene with different illuminations. Furthermore, we improve the results by incorporating a bilateral solver layer into our system during both training and test stages. Experimental results show that our approach produces better results than the state-of-the-art DL and non-DL methods on various synthetic and real datasets both visually and numerically.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信