深圳大学：《图片处理基础》课程教学课件（PPT讲稿）Poisson Image Editing

塚圳大学可视计算研究中心 SIIENZHIEN UNIVERSITY Poisson Image Editing Patric Perez, michel gangnet and Andrew Black (SIGGRAPH 2003) Presentation by Lingli He 6,jun2017

Poisson Image Editing 1 Patric Perez, Michel Gangnet, and Andrew Black (SIGGRAPH 2003) Presentation by : Lingli He 6 , jun 2017

Input Images source Image target image cc可视计算研究中心 Visual Computing Research Center

Input Images 2 source image target image

Editing Results 个 Simple cloning result Poisson Seamless Cloning cc可视计算研究中心

Editing Results 3 Simple Cloning Result Poisson Seamless Cloning

Motivation source images target image simple cloning Poisson seamless cioning Problems how to remove the seams between the mixed images cc可视计算研究中心 Visual Computing Research Center

Motivation 4 • Problems :how to remove the seams between the mixed images

Goals importing(cloning)transparent and opaque source image regions into a destination image in a seamless and effortless manner Seamless modification of appearance of the image within a selected region cc可视计算研究中心 Visual Computing Research Center

• importing (cloning) transparent and opaque source image regions into a destination image in a seamless and effortless manner. • Seamless modification of appearance of the image within a selected region. Goals 5

Background a famous psychologists show that the second order variations extracted by the Laplacian operator are the most significant perceptually (Land and McCann 1971) a scalar function on a bounded domain is uniquely defined by its values on the boundary and its Laplacian in the interior The Poisson equation has a unique solution vcc可视计算研究中心

• A famous psychologists show that the second￾order variations extracted by the Laplacian operator are the most significant perceptually (Land and McCann 1971) • A scalar function on a bounded domain is uniquely defined by its values on the boundary and its Laplacian in the interior. • The Poisson equation has a unique solution. Background 6

Related work- poisson equation Fattalet al. 2002 Entire images Lewis 2001 harmonic interpolation Our method Local images Our method Guided Interpolation cc可视计算研究中心 Visual Computing Research Center

Related work– poisson equation 7 Fattal et al. 2002 Entire images Our method Local images Lewis 2001 harmonic interpolation Our method Guided Interpolation

Related work--multiresolution images blending Burt and Adelson 1983 long range (larger gradient) mixing Poisson seamless cloning small gradient source/destination color transfer monochrome transfer cc可视计算研究中心 Visual Computing Research Center

• Burt and Adelson1983 long range(larger gradient) mixing • Poisson seamless cloning small gradient Related work—multiresolution images blending 8

Related work- seamless cloning VAL inpainting techniques-PDE. more complex EXample-based interpolation methods than poisson equation( Bertalmio et al handle large holes and textured boundaries 2000) Barret and Cheney 2002) a guided interpolation framework, with the guidance being selected by the user and this framework is not limited in seamless Cloning cc可视计算研究中心 Visual Computing Research Center

inpainting techniques-PDE . more complex than poisson equation ( Bertalmio et al. 2000 ) Related work– seamless cloning 9 a guided interpolation framework , with the guidance being selected by the user ,and this framework is not limited in seamless cloning Example-based interpolation methods ; handle large holes and textured boundaries ( Barret and Cheney 2002)

Interpolation Problem S S: a closed subset of r2 Q2: a closed subset of S, with boundary aQ fr known scalar function over sio ·f: unknown scalar function overΩ cc可视计算研究中心

• S: a closed subset of R2 • Ω: a closed subset of S, with boundary ∂Ω • f*: known scalar function over S\Ω • f: unknown scalar function over Ω Interpolation Problem 10