Intrinsic image / video

https://blog.csdn.net/frank_xu_0818/article/details/38456781

作者写于2014-8 链接如上

首先看分解图像的效果：

                                     original                                                                            reflectance

                                       shading                                                                           specularity

                                   fig.1   http://www.cs.toronto.edu/~rgrosse/intrinsic/images/panther.html

参考文献：

【1】 Ground truth dataset and baseline evaluations for intrinsic image algorithms. ICCV, 2009.

         项目主页  http://www.cs.toronto.edu/~rgrosse/intrinsic/  含有 Python  Code 及 Data

         分解表示：

                  I(x)  =  S(x)*R(x) + C(x)

          其中，I(x)  ---- pixel intensity,  S(x) ---- illuminance,  R(x) ---- albedo, C(x) ---- specular

【2】Recovering Intrinsic Images with a Global Sparsity Prior on  Reflectance.  NIPS, 2011. 

          项目主页 http://people.tuebingen.mpg.de/mkiefel/projects/intrinsic/    code： matlab/python/C++

          问题简述与模型：

     

【3】 User-assisted Intrinsic Images.  ACM SIGGRAPH Aisa ,2009.

          项目主页   http://maverick.inria.fr/Publications/2009/BPD09/index.php

          http://people.csail.mit.edu/sparis/

          https://sites.google.com/site/masterreports/nov3

          http://code.google.com/p/jh-cv/source/browse/trunk/IntrinsicImage/?r=188#IntrinsicImage

【4】 Intrinsic Images Using Optimization. CVPR, 2011

          Intrinsic images decomposition using  optimization and  user  scribbles. IEEE Trans. on Cybernetics, 2013.

          http://www.vision.ee.ethz.ch/~shenj/

【5】 A Closed-form Solution to Retinex with Non-local Texture Constraints. TPAMI, 2012.

          source code ： http://www.ece.nus.edu.sg/stfpage/eletp/SourceCode.zip    C++

【6】 SIRFS 系列

          Color Constancy, Intrinsic Images, and Shape Estimation. ECCV, 2012.

          Intrinsic Sence Properties from a single RGB-D Image.  CVPR, 2013.

          Shape, Illumination, and Reflection from Shading. Tech Report, 2013. 

          详见：Jon Barron 个人主页  http://www.cs.berkeley.edu/~barron/

【7】 Exploiting Reflection Change for Automatic Reflection Removal. ICCV, 2013.

          source code ： http://www.comp.nus.edu.sg/~liyu1988/paper/iccv2013/Reflection_Removal_Code_v1.1.zip

          另有文章：A Computational Approach for Obstruction-Free Photography http://people.csail.mit.edu/celiu/pdfs/ObstructionFree.pdf

                     

【8】 Single Image Layer Separation using Relative Smoothnes.  CVPR, 2014. 

          http://www.comp.nus.edu.sg/~liyu1988/paper/cvpr2014/cvpr2014.pdf

                  

【9】 Intrinsic Video.  ECCV, 2014.    http://ps.is.tue.mpg.de/project/Intrinsic_Video

    

【10】  Bayesian Nonparametric Intrinsic Image Decomposition. ECCV, 2014.

             链接地址：  http://people.csail.mit.edu/jchang7/pubs/publications/chang14_ECCV.pdf

【11】 Intrinsic Video and Applications.  SIGGRAPH, 2014.

            项目主页 http://media.au.tsinghua.edu.cn/yegenzhi/IntrinsicVideo.htm

【12】Intrinsic Images in the Wild. SIGGRAPH, 2014.  复杂场景的 reflectance and shading 图像分解

           项目主页  http://opensurfaces.cs.cornell.edu/intrinsic/

【13】2015 ICCV 

     A Comprehensive Multi-illuminant Dataset for Benchmarking of the Intrinsic Image Algorithms  http://www.cg.informatik.uni-siegen.de/en/shida

     Learning Data-driven Reflectance Priors for Intrinsic Image Decomposition

   Intrinsic Depth: Improving Depth Transfer with Intrinsic Images  https://ps.is.tuebingen.mpg.de/publications/-50251d51-15d1-4463-b4d4-8423b03c9a95

   Direct Intrinsics: Albedo and Shading Decomposition by CNN Regression