Image Super-Resolution by Neural Texture Transfer(CVPR2019)
Motivation：基於參考圖像的圖像超分方法具有很大的潛力，但是當參考圖像和LR圖像不夠相似時，或者未對齊時，圖像的超分效果就會很差。 整體思路： 本文旨在利用來之參考圖像的更多紋理信息來釋放RefSR的潛力，即是在不提供相關的Ref圖象時，也具有很強的魯棒性。將RefSR問題形式化爲神經紋理遷移問題，設計了端到端的模型，將Ref圖像的紋理自適應的遷移，從而豐富HR圖像細節，且在神經空間執行多
相關文章
- 1. Image Super-Resolution by Neural Texture Transfer
- 2. 讀論文：Image Super-Resolution by Neural Texture Transfer
- 3. [論文筆記] Image Super-Resolution by Neural Texture Transfer (CVPR2019)
- 4. Stereoscopic image quality assessment by deep convolutional neural network
- 5. Recovering Realistic Texture in Image Super-resolution by Deep Spatial Feature Transform
- 6. Laya中的Image、Texture、WebGLImage
- 7. NIMA: Neural Image Assessment
- 8. High-Resolution Image Inpainting using Multi-Scale Neural Patch Synthesis
- 9. Reading Note: Neural Aesthetic Image Reviewer
- 10. Stereo Matching by Training a Convolutional Neural Network to Compare Image Patches 譯文
- 更多相關文章...
- • RSS
元素 - RSS 教程 - • ASP.NET Image 控件 - ASP.NET 教程
- • Docker 清理命令
- • RxJava操作符（一）Creating Observables
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. Image Super-Resolution by Neural Texture Transfer
- 2. 讀論文：Image Super-Resolution by Neural Texture Transfer
- 3. [論文筆記] Image Super-Resolution by Neural Texture Transfer (CVPR2019)
- 4. Stereoscopic image quality assessment by deep convolutional neural network
- 5. Recovering Realistic Texture in Image Super-resolution by Deep Spatial Feature Transform
- 6. Laya中的Image、Texture、WebGLImage
- 7. NIMA: Neural Image Assessment
- 8. High-Resolution Image Inpainting using Multi-Scale Neural Patch Synthesis
- 9. Reading Note: Neural Aesthetic Image Reviewer
- 10. Stereo Matching by Training a Convolutional Neural Network to Compare Image Patches 譯文