Free Appearance-Editing with Improved Poisson Image Cloning
|
|
- Brittany Taylor
- 6 years ago
- Views:
Transcription
1 Bie XH, Huang HD, Wang WC. Free appearance-editing with improved Poisson image cloning. JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY 26(6): Nov DOI /s Free Appearance-Editing with Improved Poisson Image Cloning Xiao-Hui Bie 1,3 ( ), Hao-Da Huang 2 ( ), and Wen-Cheng Wang 1 ( ), Member, CCF 1 State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences, Beijing , China 2 Microsoft Research Asia, Beijing , China 3 Graduate University of Chinese Academy of Sciences, Beijing , China xiaohui@ios.ac.cn; hahuang@microsoft.com; whn@ios.ac.cn Received December 29, 2010; revised September 5, Abstract In this paper, we present a new edit tool for the user to conveniently preserve or freely edit the object appearance during seamless image composition. We observe that though Poisson image editing is effective for seamless image composition. Its color bleeding (the color of the target image is propagated into the source image) is not always desired in applications, and it provides no way to allow the user to edit the appearance of the source image. To make it more flexible and practical, we introduce new energy terms to control the appearance change, and integrate them into the Poisson image editing framework. The new energy function could still be realized using efficient sparse linear solvers, and the user can interactively refine the constraints. With the new tool, the user can enjoy not only seamless image composition, but also the flexibility to preserve or manipulate the appearance of the source image at the same time. This provides more potential for creating new images. Experimental results demonstrate the effectiveness of our new edit tool, with similar time cost to the original Poisson image editing. Keywords poisson image editing, appearance editing, edit propagation 1 Introduction Image composition is popular in creating new images by pasting an object or a region from a source image onto a target image. For seamless image composition, Poisson image editing [1] has been demonstrated as a very effective approach. By solving Poisson equations using the user-specified boundary condition, it can have the colors along the boundary between the source and target images changed slowly, and so blend the colors from both images without visible discontinuities around the boundary. However, such a reconstruction function interpolates the boundary conditions inwards, so the appearance of the objects in the source image would change, called color bleeding. In applications, such a color change is not always desired, because the interesting contents of the source image may be smudged to violate the desire of image composition, as illustrated in Fig.1, where the dog of the source image has its color changed in a miscellany of white and carnation with Poisson image editing. This may seriously prevent the use of Poisson image editing. For the color bleeding problem in Poisson image editing, we have not found efficient solutions in literatures till now, though it was listed as one of the future research directions in 2006 [2]. In this paper, we propose a new edit tool to well handle the color bleeding in Poisson image cloning. It cannot only well preserve the object appearance in the source image, but also allow the user to edit the object appearance freely. At first, it lets the user use strokes to specify the regions that will have their appearance preserved or edited. Then, a new energy function is designed for Poisson image editing by introducing two new energy terms, one for responding to the user s desire and the other for propagating the user s desire to the regions similar to the specified regions. In this way, the regions of interest can resist the influence from the target image so that their appearance can be preserved or edited freely. The new energy function can be still solved using the efficient sparse linear solver as in [1]. Our new tool can perform image composition quickly to soon respond to the user s interactions, and so is very helpful for the user to produce his desired images by adding constraints progressively. As shown in Fig.1, with our new edit tool, the dog can have its white color preserved or freely edited to be other pure colors. Therefore, our new edit tool provides more potential for the user to create new images with image composition. Short Paper This work was partly supported by the National Natural Science Foundation of China under Grant Nos , , Springer Science + Business Media, LLC & Science Press, China
2 1012 J. Comput. Sci. & Technol., Nov. 2011, Vol.26, No.6 Fig.1. Free appearance-editing in Poisson image cloning. Given a source image (a), it is pasted onto a target image (b), where the user draws strokes in orange to specify the dog to have its color preserved or edited. With Poisson image editing, the dog has its color changed to be in a miscellany of white and carnation (c), due to color bleeding. With our new method, the dog can retain its color (d), or be edited in yellow (e) or green (f) freely. 2 Related Work 2.1 Image Composition There are many approaches for image composition. By image matting, existing methods always require the user to provide additional constraints using a trimap [3] or a set of brush strokes [4], and some techniques [5] employ a controlled environment or a special device to reduce the inherent ambiguity of the matting problem. For seamless blending of the target and source images, the graph-cut algorithm has been adopted to find the best seams between the combined regions [6-7], and Poisson image editing is generally regarded as a more efficient technique [1]. When the source and target images are taken from different sources or shot under different conditions, simply using the above techniques may result in heavy inconsistencies. By using a multi-scale image harmonization method before composition [8], such inconsistencies can be successfully removed and more realistic composites will be produced. To make Poisson image editing more practical, many improvements have been made. In one aspect, many acceleration techniques have been proposed [9-12]. They all work very well as long as the boundary is suitable and the interior is smooth. In another aspect, boundary conditions are studied a lot. For example, Poisson matting [3] tries to efficiently treat blurry objects, and Drag-and-Drop pasting [2] enables the users to easily select the regions of interest by automatically reducing the conflict between the salient structures in the source and target images along the boundary. But none of them addresses the color bleeding problem in Poisson image editing. In comparison with these techniques, our new technique can efficiently support appearance preserving in Poisson image editing, and even appearance editing in seamless image composition. Experiments show that it is valid for a wide variety of images and able to run much quickly. This provides an efficient solution to the color bleeding problem in Poisson image editing. 2.2 Appearance Editing It is important to adjust the image appearance in image editing. To perform such adjustments in a particular region, e.g., increasing or decreasing the exposure locally, the user need first carefully select the region. To simplify the region selection process, several intuitive stroke-based image editing tools have been developed. In general, first the user marks the region of interest with a few strokes, and specifies constraints on the marked regions, then propagates the constraints to the entire image by solving an optimization problem. These methods have been proved effective for images [13-14] as well as materials [15-16]. However, they are generally limited to editing a single image or a kind of material, not addressing the editing problem in pasting a source image onto a target image. As for our new method, it also utilizes strokes for the user to specify the region of interest, and gives constraints to propagate for appearance preserving or editing. However, it tries to composite two images with appearance editing and image composition solved simultaneously in an unified framework for Poisson image cloning. Recently, Farbman et al. [17] proposed to use diffusion distances instead of generally used Euclidean distances for local edit propagation, because diffusion distances are more efficient to make edge information aware in propagating edits and helpful to promote edited results. It could be used for color composition, but, it needs two steps with the first to paste the source image onto the target image, and the second to propagate edits, where a suitable diffusion map has to be produced on the fly by setting good parameters according to the result of the first step. In comparison with this work,
3 Xiao-Hui Bie et al.: Free Appearance Editing in Image Cloning 1013 our method needs only one step without the expensive computation of the diffusion map. Thus, our method would be more convenient to composite images. 3 Free Appearance-Editing in Poisson Image Cloning We address the color bleeding problem in Poisson image editing in this section and try to propose an interactive tool for appearance editing in image composition. Using a few strokes, the user can easily specify the regions where he wants to keep or change the appearance. Then, the constraints are considered to form a new energy function to be minimized in the framework of Poisson image editing. As a result, the source image can be seamlessly blended into the target image while the specified regions can still retain their appearance or be changed as desired. 3.1 Poisson Image Editing In Poisson image editing [1], the membrane interpolant for blending the source image g and the target image f is defined as the solution of the minimization problem: min f v 2 dp with f Ω = f Ω, (1) p Ω where f is the resulting image, v = g is the guidance field for seamless cloning, Ω is the domain covered by the source image and Ω is the exterior boundary of Ω. These notations are illustrated in Fig.2. Fig.2. Guided interpolation notations [1]. Unknown function f interpolates the destination function f in domain Ω, under guidance of vector v, which might be or not be the gradient field of a source function g. For discrete images the above problem can be discretized using the underlying discrete pixel grids. Without ambiguity, suppose that S and Ω represent the pixels of the target and the source images respectively. For each pixel p in S, let N p be the set of its 4-connected neighbors which are in S, and let p, q denote a pixel pair such that q N p. The boundary of Ω is now Ω = {p S \ Ω : N p Ω }. Let f p be the value of f at p, the finite difference discretization of (1) yields the following discrete, quadratic optimization problem: for all p Ω, (f p f q v pq ) 2, with f p = fp, (2) min f Ω p,q Ω where v pq is the projection of v((p + q)/2) on the oriented edge [p, q], i.e., v pq = v((p + q)/2) pq. Its solution satisfies the following simultaneous linear equations: N p f p f q = fq + v pq, q N p Ω q N p q N p Ω for all p Ω. (3) As (3) forms a classical, sparse (banded), symmetric, and positive-definite system, it can be solved fast enough for interactive editing of medium-sized color image regions. 3.2 New Energy Function For appearance editing in Poisson image cloning, we put forward two energy terms to add to the (2), so as to form a new minimized energy function as follows: E = (f p f q v pq ) 2 + p,q Ω ( α ((f p g p ) e p ) 2 + p M β ) ((f p g p ) (f q g q )) 2 z pq, p Ω q N p (4) where α and β are two weights for the user to decide how to edit the appearance of the specified regions, M is the specified regions in the source image for appearance editing, g p represents the value at p in the original source image patch, e p represents the change that the user wants to assign at p, and z pq is for measuring the color similarity between the pixels p and q, computed in exp( ((R p R q ) 2 + (G p G q ) 2 + (B p B q ) 2 )/2σ 2 ) with σ being the variance of the color values at the pixels inside Ω. Obviously, when α = 0, it returns to the original Poisson image editing. For the term ((f p g p ) e p ) 2, it tries to retain the original value at p when e p = 0, or have the change at p aligned with e p. As for the term ((f p g p ) (f q g q )) 2 z pq, it is for propagating the editing constraints given by the user to the regions similar to the specified regions in Ω. By the new function, the solution would satisfy the following simultaneous linear equation: ( N p + α + αβ z pq )f p (1 + αβz pq )f q q N p q N p Ω
4 1014 J. Comput. Sci. & Technol., Nov. 2011, Vol.26, No.6 = q N p Ω (1 + αβz pq )f q + q N p (1 + αβz pq )v pq + αg p, for all p Ω. (5) In this way, the user s desire can be integrated into Poisson image editing for appearance preserving or manipulating in seamless image composition. 4 Results and Discussion With our new energy function, appearance editing can be easily implemented in Poisson image cloning. The related interpolations are also solved in Gauss- Seidel iteration with successive overrelaxation, as suggested in [1]. With the new tool, we conducted experiment on a wide variety of source and target images, and compared it with Poisson image editing. In the tests, α and β are set to be 1, 10 or 100 respectively. In fact, our new method is insensitive to their values by a lot of tests. By our tests on a personnel computer with an Intel Q9400 CPU and 4 G RAM, our new method ran at a similar speed to the original Poisson image editing. For example, in handling the examples Dog, Kid and Flower, our method took 0.29 s, 0.73 s and 0.25 s respectively for image composition, while our implementation Fig.3. Kid. (a) Source image. (b) Target image. (c) When the source image is pasted onto the target image, the kid is specified in strokes. (d) The membrane map to show the color change of the pixels covered by the source image patch after our new method is applied. (e) The result with Poisson image editing where the kid has her appearance changed a lot by the target image, due to color bleeding. (f) Our result where the appearance of the kid in the source image is well preserved. of original Poisson editing took 0.36 s, 0.67 s and 0.31 s respectively. Appearance Preserving. If the source and target images differ a lot in color and structure, the source image would be smudged with Poisson image editing, as shown in Fig.3(e). But with our new edit tool, the regions of interest can preserve the appearance shown in Fig.3(f) after they are specified in strokes in Fig.3(c). In Fig.3(d), the membrane map is given to show the color change in the source image with our new method, where the colors correspond to the differences in R, G and B at the pixels respectively. It is obvious that the specified regions have little or no change. This demonstrates that our new method can efficiently preserve appearance in Poisson image cloning. Appearance Editing. Because our new method can make the regions of interest resist the influence from the target image, forming independent areas irrespective of the blending of the source and target images, the user can freely edit the appearance of the regions of interest, as shown in Fig.1. As discussed in [2], the source image of fog plus a chimney is not easy to handle image matting. With such an image, our new method can easily preserve the appearance of the white fog, as shown in Fig.4(d), or edit the fog in other colors in Figs. 4(e) 4(g), respectively. Selective Preservation with Seamless Blending. Not all the color bleeding effects are disadvantageous to image composition. Sometimes, color bleeding effects by Poisson image editing are very helpful to create a photo realistic composite. With our method, we can selectively maintain the favorable color bleeding effects while remove the unfavorable ones by the user s intention. However, this is unable to achieve with image matting, since image matting cannot obtain color bleeding effects. As illustrated in Fig.5, our method is able to maintain the color bleeding effects under the water to achieve a seamless composition, while preserves the original appearance of the boy s head in the source image. But for matting, as illustrated in Figs. 5(e) 5(h), it is difficult to find a suitable matte to produce a photo realistic composite. Limitation. By our new method, the constraints that the user gives to the regions of interest will be propagated to the similar regions. When the regions of interest are similar to other regions in color, the edit constraints may be propagated to other regions to make the composite inconsistent along the boundary. 5 Conclusions Color bleeding in Poisson image cloning is not always desired in applications, which prevents the wide use of Poisson image editing. With regard to this, this
5 Xiao-Hui Bie et al.: Free Appearance Editing in Image Cloning 1015 Fig.4. Fog. (a) Source image. (b) The fog is specified in strokes when the source image patch is pasted onto the target image. (c) The result with Poisson image editing, where the fog has its color changed due to color bleeding. (d) Our result with appearance editing, where the fog remains its color in the source image. With our new method, the fog can be freely edited in various colors, as shown in (e) (g) respectively. Fig.5. Swimming boy. (a) Source image. (b) The boy s head is specified in strokes after the source image patch is pasted onto the target image. (c) The result with Poisson image editing, where the boy s head has its color changed due to color bleeding. (d) Our result with the appearance of the boy s head well preserved, while the boy s body has the appearance changed to achieve a seamless composition. (e) The alpha matte computed by the matting method [4] when we specify the boy s head, breast and left hand as foreground. (f) The result with the alpha matte obtained in (e), where much of the boy s body under the water cannot be seen. (g) Another alpha matte when we specify all the boy s body as foreground. (h) The result with the alpha matte obtained in (g), where the boundary of the matte is obvious and the source and target images are not blended seamlessly. paper proposed two new energy terms for Poisson image editing to well preserve or even freely edit the appearance of the regions of interest in the source image, while composing images seamlessly. Thus, it provides more potential for the user to create new images. This is a general framework. It can not only freely edit appearance in image composition, but also easily return to the original Poisson image editing when required. Experimental results have demonstrated its effectiveness. References [1] Pérez P, Gangnet M, Blake A. Poisson image editing. ACM Trans. Graph, 2003, 22(3): [2] Jia J, Sun J, Tang C K, Shum H Y. Drag-and-drop pasting. ACM Trans. Graph, 2006, 25(3):
6 1016 J. Comput. Sci. & Technol., Nov. 2011, Vol.26, No.6 [3] Sun J, Jia J, Tang C K, Shum H Y. Poisson matting. ACM Trans. Graph, 2004, 23(3): [4] Levin A, Rva-acha A, Lischinski D. Spectral matting. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 30(10): [5] Mcguire M, Matusik W, Pfister H, Hughes J F, Durand F. Defocus video matting. ACM Trans. Graph, 2005, 24(3): [6] Boykov Y, Jolly M. Interactive graph cuts for optimal boundary and region segmentation of objects in N-D images. In Proc. ICCV 2001, Vancouver, Canada, July 7-14, 2001, pp [7] Agarwala A, Dontcheva M, Agrawala M, Drucker S, Colburn A, Curless B, Salesin D, Cohen M. Interactive digital photomontage. ACM Trans. Graph, 2004, 23(3): [8] Sunkavalli K, Johson M K, Matusik W, Pfister H. Multi-scale image harmonization. ACM Trans. Graph, 2010, 29(4). [9] Kazhdan M, Hoppe H. Streaming multigrid for gradientdomain operations on large images. ACM Trans. Graph, 2008, 27(3). [10] Agarwala A. Efficient gradient-domain compositing using quadtrees. ACM Trans. Graph, 2007, 27(3). [11] Farbman Z, Hoffer G, Lipman Y, Cohen-Or D, Lichinski D. Coordinates for instant image cloning. ACM Trans. Graph, 2009, 28(3). [12] Jeschke S, Cline D, Wonka P. A GPU laplacian solver for diffusion curves and poisson image editing. ACM Trans. Graph, 2009, 28(5). [13] Levin A, Lischinski D, Weiss Y. Colorization using optimization. ACM Trans. Graph, 2004, 23(3): [14] Lischinshi D, Farbman Z, Uyttendaele M, Szeliski R. Interactive local adjustment of tonal values. ACM Trans. Graph, 2006, 25(3): [15] Pellacini F, Lawrence J. Appwand: Editing measured materials using appearance-driven optimization. ACM Trans. Graph. 2007, 26(3). [16] An X, Pellacini F. Appprop: All-pairs appearance-space edit propagation. ACM Trans. Graph, 2008, 27(3). [17] Farbman Z, Fattal R, Lischinski D. Diffusion maps for edgeaware image editing. ACM Trans. Graph, 2010, 29(6). Xiao-Bui Bie received his B.S. degree in engineering mechanics from Huazhong University of Science and Technology. He is currently a Ph.D. student at the Institute of Software, Chinese Academy of Sciences. His main research interests are computer vision and interactive image/video editing. Hao-Da Huang received his Master s degree from the Institute of Software, Chinese Academy of Sciences in He is currently an associate researcher of Microsoft Research Asia. His research interests include facial performance capture, hand deformation, image analysis and editing. Wen-Cheng Wang received his Ph.D. degree from the Institute of Software, Chinese Academy of Sciences in 1998, where he is currently a professor of the State Key Laboratory of Computer Science. His research interests include computer graphics, visualization, virtual reality and expressive rendering and editing.
Color Me Right Seamless Image Compositing
Color Me Right Seamless Image Compositing Dong Guo and Terence Sim School of Computing National University of Singapore Singapore, 117417 Abstract. This paper introduces an approach of creating an image
More informationIntent-aware image cloning
Vis Comput DOI 10.1007/s00371-013-0826-0 ORIGINAL ARTICLE Intent-aware image cloning Xiaohui Bie Wencheng Wang Hanqiu Sun Haoda Huang Minying Zhang Springer-Verlag Berlin Heidelberg 2013 Abstract Currently,
More informationFast Image Stitching and Editing for Panorama Painting on Mobile Phones
Fast Image Stitching and Editing for Panorama Painting on Mobile Phones Yingen Xiong and Kari Pulli Nokia Research Center 955 Page Mill Road, Palo Alto, CA 94304, USA {yingen.xiong, kari.pulli}@nokia.com
More informationFast Image Stitching and Editing for Panorama Painting on Mobile Phones
in IEEE Workshop on Mobile Vision, in Conjunction with CVPR 2010 (IWMV2010), San Francisco, 2010, IEEE Computer Society Fast Image Stitching and Editing for Panorama Painting on Mobile Phones Yingen Xiong
More informationDrag and Drop Pasting
Drag and Drop Pasting Jiaya Jia, Jian Sun, Chi-Keung Tang, Heung-Yeung Shum The Chinese University of Hong Kong Microsoft Research Asia The Hong Kong University of Science and Technology Presented By Bhaskar
More informationFast Image Labeling for Creating High-Resolution Panoramic Images on Mobile Devices
Multimedia, IEEE International Symposium on, vol. 0, pp. 369 376, 2009. Fast Image Labeling for Creating High-Resolution Panoramic Images on Mobile Devices Yingen Xiong and Kari Pulli Nokia Research Center
More informationGradient Domain Image Blending and Implementation on Mobile Devices
in MobiCase 09: Proceedings of The First Annual International Conference on Mobile Computing, Applications, and Services. 2009, Springer Berlin / Heidelberg. Gradient Domain Image Blending and Implementation
More informationIMPLEMENTATION OF THE CONTRAST ENHANCEMENT AND WEIGHTED GUIDED IMAGE FILTERING ALGORITHM FOR EDGE PRESERVATION FOR BETTER PERCEPTION
IMPLEMENTATION OF THE CONTRAST ENHANCEMENT AND WEIGHTED GUIDED IMAGE FILTERING ALGORITHM FOR EDGE PRESERVATION FOR BETTER PERCEPTION Chiruvella Suresh Assistant professor, Department of Electronics & Communication
More informationPhotoshop Quickselect & Interactive Digital Photomontage
Photoshop Quickselect & Interactive Digital Photomontage By Joseph Tighe 1 Photoshop Quickselect Based on the graph cut technology discussed Boykov-Kolmogorov What might happen when we use a color model?
More informationColor Adjustment for Seamless Cloning based on Laplacian-Membrane Modulation
Color Adjustment for Seamless Cloning based on Laplacian-Membrane Modulation Bernardo Henz, Frederico A. Limberger, Manuel M. Oliveira Instituto de Informática UFRGS Porto Alegre, Brazil {bhenz,falimberger,oliveira}@inf.ufrgs.br
More informationIntroduction to Computer Graphics. Image Processing (1) June 8, 2017 Kenshi Takayama
Introduction to Computer Graphics Image Processing (1) June 8, 2017 Kenshi Takayama Today s topics Edge-aware image processing Gradient-domain image processing 2 Image smoothing using Gaussian Filter Smoothness
More informationImage Stitching using Watersheds and Graph Cuts
Image Stitching using Watersheds and Graph Cuts Patrik Nyman Centre for Mathematical Sciences, Lund University, Sweden patnym@maths.lth.se 1. Introduction Image stitching is commonly used in many different
More informationAVOIDING BLEEDING IN IMAGE BLENDING. TNList, Tsinghua University School of Computer Science and Informatics, Cardiff University
AVOIDING BLEEDING IN IMAGE BLENDING Minxuan Wang Zhe Zhu Songhai Zhang Ralph Martin Shi-Min Hu TNList, Tsinghua University School of Computer Science and Informatics, Cardiff University ABSTRACT Though
More informationProf. Feng Liu. Spring /17/2017. With slides by F. Durand, Y.Y. Chuang, R. Raskar, and C.
Prof. Feng Liu Spring 2017 http://www.cs.pdx.edu/~fliu/courses/cs510/ 05/17/2017 With slides by F. Durand, Y.Y. Chuang, R. Raskar, and C. Rother Last Time Image segmentation Normalized cut and segmentation
More informationImage Super-Resolution by Vectorizing Edges
Image Super-Resolution by Vectorizing Edges Chia-Jung Hung Chun-Kai Huang Bing-Yu Chen National Taiwan University {ffantasy1999, chinkyell}@cmlab.csie.ntu.edu.tw robin@ntu.edu.tw Abstract. As the resolution
More informationSoft Scissors : An Interactive Tool for Realtime High Quality Matting
Soft Scissors : An Interactive Tool for Realtime High Quality Matting Jue Wang University of Washington Maneesh Agrawala University of California, Berkeley Michael F. Cohen Microsoft Research Figure 1:
More informationIMA Preprint Series # 2153
DISTANCECUT: INTERACTIVE REAL-TIME SEGMENTATION AND MATTING OF IMAGES AND VIDEOS By Xue Bai and Guillermo Sapiro IMA Preprint Series # 2153 ( January 2007 ) INSTITUTE FOR MATHEMATICS AND ITS APPLICATIONS
More informationThis work is about a new method for generating diffusion curve style images. Although this topic is dealing with non-photorealistic rendering, as you
This work is about a new method for generating diffusion curve style images. Although this topic is dealing with non-photorealistic rendering, as you will see our underlying solution is based on two-dimensional
More informationVideo Operations in the Gradient Domain. Abstract. these operations on video in the gradient domain. Our approach consists of 3D graph cut computation
Video Operations in the Gradient Domain 1 Abstract Fusion of image sequences is a fundamental operation in numerous video applications and usually consists of segmentation, matting and compositing. We
More information3D Editing System for Captured Real Scenes
3D Editing System for Captured Real Scenes Inwoo Ha, Yong Beom Lee and James D.K. Kim Samsung Advanced Institute of Technology, Youngin, South Korea E-mail: {iw.ha, leey, jamesdk.kim}@samsung.com Tel:
More informationAssignment 4: Seamless Editing
Assignment 4: Seamless Editing - EE Affiliate I. INTRODUCTION This assignment discusses and eventually implements the techniques of seamless cloning as detailed in the research paper [1]. First, a summary
More informationAdding a Transparent Object on Image
Adding a Transparent Object on Image Liliana, Meliana Luwuk, Djoni Haryadi Setiabudi Informatics Department, Petra Christian University, Surabaya, Indonesia lilian@petra.ac.id, m26409027@john.petra.ac.id,
More informationImage-Based Rendering for Ink Painting
2013 IEEE International Conference on Systems, Man, and Cybernetics Image-Based Rendering for Ink Painting Lingyu Liang School of Electronic and Information Engineering South China University of Technology
More informationEDGE-AWARE IMAGE PROCESSING WITH A LAPLACIAN PYRAMID BY USING CASCADE PIECEWISE LINEAR PROCESSING
EDGE-AWARE IMAGE PROCESSING WITH A LAPLACIAN PYRAMID BY USING CASCADE PIECEWISE LINEAR PROCESSING 1 Chien-Ming Lu ( 呂建明 ), 1 Sheng-Jie Yang ( 楊勝傑 ), 1 Chiou-Shann Fuh ( 傅楸善 ) Graduate Institute of Computer
More informationIMA Preprint Series # 2171
A GEODESIC FRAMEWORK FOR FAST INTERACTIVE IMAGE AND VIDEO SEGMENTATION AND MATTING By Xue Bai and Guillermo Sapiro IMA Preprint Series # 2171 ( August 2007 ) INSTITUTE FOR MATHEMATICS AND ITS APPLICATIONS
More informationMotion Regularization for Matting Motion Blurred Objects
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 33, NO. 11, NOVEMBER 2011 2329 Motion Regularization for Matting Motion Blurred Objects Hai Ting Lin, Yu-Wing Tai, and Michael S. Brown
More informationColor retargeting: interactive time-varying color image composition from time-lapse sequences
Computational Visual Media DOI 10.1007/s41095-xxx-xxxx-x Vol. x, No. x, month year, xx xx Research Article Color retargeting: interactive time-varying color image composition from time-lapse sequences
More informationA Ray Tracing Approach to Diffusion Curves
Eurographics Symposium on Rendering 2011 Ravi Ramamoorthi and Erik Reinhard (Guest Editors) Volume 30 (2011), Number 4 A Ray Tracing Approach to Diffusion Curves John C. Bowers Jonathan Leahey Rui Wang
More informationColor Source Separation for Enhanced Pixel Manipulations MSR-TR
Color Source Separation for Enhanced Pixel Manipulations MSR-TR-2-98 C. Lawrence Zitnick Microsoft Research larryz@microsoft.com Devi Parikh Toyota Technological Institute, Chicago (TTIC) dparikh@ttic.edu
More informationAutomated Segmentation Using a Fast Implementation of the Chan-Vese Models
Automated Segmentation Using a Fast Implementation of the Chan-Vese Models Huan Xu, and Xiao-Feng Wang,,3 Intelligent Computation Lab, Hefei Institute of Intelligent Machines, Chinese Academy of Science,
More informationFast Poisson Blending using Multi-Splines
Fast Poisson Blending using Multi-Splines Richard Szeliski, Matt Uyttendaele, and Drew Steedly Microsoft Research April 2008 Technical Report MSR-TR-2008-58 We present a technique for fast Poisson blending
More informationStereo Matching on Objects with Fractional Boundary
Stereo Matching on Objects with Fractional Boundary Wei Xiong and iaya ia Department of Computer Science and Engineering The Chinese University of Hong Kong {wxiong, leojia}@cse.cuhk.edu.hk Abstract Conventional
More informationIMAGE stitching is a common practice in the generation of
IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 15, NO. 4, APRIL 2006 969 Seamless Image Stitching by Minimizing False Edges Assaf Zomet, Anat Levin, Shmuel Peleg, and Yair Weiss Abstract Various applications
More informationShift-Map Image Editing
Shift-Map Image Editing Yael Pritch Eitam Kav-Venaki Shmuel Peleg School of Computer Science and Engineering The Hebrew University of Jerusalem 91904 Jerusalem, Israel Abstract Geometric rearrangement
More informationDivide and Conquer: A Self-Adaptive Approach for High-Resolution Image Matting
Divide and Conquer: A Self-Adaptive Approach for High-Resolution Image Matting Guangying Cao,Jianwei Li, Xiaowu Chen State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science
More informationRecap. DoF Constraint Solver. translation. affine. homography. 3D rotation
Image Blending Recap DoF Constraint Solver translation affine homography 3D rotation Recap DoF Constraint Solver translation 2 affine homography 3D rotation Recap DoF Constraint Solver translation 2 affine
More informationColor retargeting: Interactive time-varying color image composition from time-lapse sequences
Computational Visual Media DOI 10.1007/s41095-015-0031-3 Vol. 1, No. 4, December 2015, 321 330 Research Article Color retargeting: Interactive time-varying color image composition from time-lapse sequences
More informationSketch Based Image Deformation
Sketch Based Image Deformation Mathias Eitz Olga Sorkine Marc Alexa TU Berlin Email: {eitz,sorkine,marc}@cs.tu-berlin.de Abstract We present an image editing tool that allows to deform and composite image
More informationAutomatic Trimap Generation for Digital Image Matting
Automatic Trimap Generation for Digital Image Matting Chang-Lin Hsieh and Ming-Sui Lee Department of Computer Science and Information Engineering, National Taiwan University, Taipei, Taiwan, R.O.C. E-mail:
More informationarxiv: v1 [cs.cv] 23 Aug 2017
Single Reference Image based Scene Relighting via Material Guided Filtering Xin Jin a, Yannan Li a, Ningning Liu c, Xiaodong Li a,, Xianggang Jiang a, Chaoen Xiao b, Shiming Ge d, arxiv:1708.07066v1 [cs.cv]
More informationPattern Recognition Letters
Pattern Recognition Letters 33 (2012) 920 933 Contents lists available at ScienceDirect Pattern Recognition Letters journal homepage: www.elsevier.com/locate/patrec Dynamic curve color model for image
More informationFOREGROUND SEGMENTATION BASED ON MULTI-RESOLUTION AND MATTING
FOREGROUND SEGMENTATION BASED ON MULTI-RESOLUTION AND MATTING Xintong Yu 1,2, Xiaohan Liu 1,2, Yisong Chen 1 1 Graphics Laboratory, EECS Department, Peking University 2 Beijing University of Posts and
More informationSINGLE UNDERWATER IMAGE ENHANCEMENT USING DEPTH ESTIMATION BASED ON BLURRINESS. Yan-Tsung Peng, Xiangyun Zhao and Pamela C. Cosman
SINGLE UNDERWATER IMAGE ENHANCEMENT USING DEPTH ESTIMATION BASED ON BLURRINESS Yan-Tsung Peng, Xiangyun Zhao and Pamela C. Cosman Department of Electrical and Computer Engineering, University of California,
More informationPaint Selection. Jian Sun Microsoft Research Asia
Paint Selection Jiangyu Liu University of Science and Technology of China Jian Sun Microsoft Research Asia Heung-Yeung Shum Microsoft Corporation Figure 1: Left three: the user makes a selection by painting
More informationImage Segmentation Using Iterated Graph Cuts Based on Multi-scale Smoothing
Image Segmentation Using Iterated Graph Cuts Based on Multi-scale Smoothing Tomoyuki Nagahashi 1, Hironobu Fujiyoshi 1, and Takeo Kanade 2 1 Dept. of Computer Science, Chubu University. Matsumoto 1200,
More informationFace Hallucination Based on Eigentransformation Learning
Advanced Science and Technology etters, pp.32-37 http://dx.doi.org/10.14257/astl.2016. Face allucination Based on Eigentransformation earning Guohua Zou School of software, East China University of Technology,
More informationFull text available at: Image and Video Matting: A Survey
Image and Video Matting: A Survey Image and Video Matting: A Survey Jue Wang Adobe Systems Incorporated 801 North 34th Street Seattle, WA 98103 USA juewang@adobe.com Michael F. Cohen Microsoft Research
More informationA New Technique for Adding Scribbles in Video Matting
www.ijcsi.org 116 A New Technique for Adding Scribbles in Video Matting Neven Galal El Gamal 1, F. E.Z. Abou-Chadi 2 and Hossam El-Din Moustafa 3 1,2,3 Department of Electronics & Communications Engineering
More informationImage Blending and Compositing NASA
Image Blending and Compositing NASA CS194: Image Manipulation & Computational Photography Alexei Efros, UC Berkeley, Fall 2016 Image Compositing Compositing Procedure 1. Extract Sprites (e.g using Intelligent
More informationParallax-tolerant Image Stitching
Parallax-tolerant Image Stitching Fan Zhang and Feng Liu Department of Computer Science Portland State University {zhangfan,fliu}@cs.pdx.edu Abstract Parallax handling is a challenging task for image stitching.
More informationFILTER BASED ALPHA MATTING FOR DEPTH IMAGE BASED RENDERING. Naoki Kodera, Norishige Fukushima and Yutaka Ishibashi
FILTER BASED ALPHA MATTING FOR DEPTH IMAGE BASED RENDERING Naoki Kodera, Norishige Fukushima and Yutaka Ishibashi Graduate School of Engineering, Nagoya Institute of Technology ABSTRACT In this paper,
More informationDetail Preserving Shape Deformation in Image Editing
Detail Preserving Shape Deformation in Image Editing Hui Fang Google, Inc. (a) John C. Hart University of Illinois, Urbana-Champaign (c) (b) (d) (e) Figure 1: The deformation of a source image (a), described
More informationPanoramic Image Stitching
Mcgill University Panoramic Image Stitching by Kai Wang Pengbo Li A report submitted in fulfillment for the COMP 558 Final project in the Faculty of Computer Science April 2013 Mcgill University Abstract
More informationBackground Estimation for a Single Omnidirectional Image Sequence Captured with a Moving Camera
[DOI: 10.2197/ipsjtcva.6.68] Express Paper Background Estimation for a Single Omnidirectional Image Sequence Captured with a Moving Camera Norihiko Kawai 1,a) Naoya Inoue 1 Tomokazu Sato 1,b) Fumio Okura
More informationCIS581: Computer Vision and Computational Photography Project 2: Face Morphing and Blending Due: Oct. 17, 2017 at 3:00 pm
CIS581: Computer Vision and Computational Photography Project 2: Face Morphing and Blending Due: Oct. 17, 2017 at 3:00 pm Instructions This is an individual project. Individual means each student must
More informationEfficient Antialiased Edit Propagation for Images and Videos
Efficient Antialiased Edit Propagation for Images and Videos Li-Qian Ma a, Kun Xu a, a TNList, Department of Computer Science and Technology, Tsinghua University, Beijing Abstract Edit propagation on images/videos
More informationRecap from Monday. Frequency domain analytical tool computational shortcut compression tool
Recap from Monday Frequency domain analytical tool computational shortcut compression tool Fourier Transform in 2d in Matlab, check out: imagesc(log(abs(fftshift(fft2(im))))); Image Blending (Szeliski
More informationEfficient Affinity-based Edit Propagation using K-D Tree
Efficient Affinity-based Edit Propagation using K-D Tree Kun Xu 1 Yong Li 1 Tao Ju 2 Shi-Min Hu 1 Tian-Qiang Liu 1 1 Tsinghua National Laboratory for Information Science and Technology and Department of
More informationPerception-based Seam-cutting for Image Stitching
Perception-based Seam-cutting for Image Stitching Nan Li Tianli Liao Chao Wang Received: xxx / Accepted: xxx Abstract Image stitching is still challenging in consumerlevel photography due to imperfect
More informationTomorrow s Photoshop Effects
Tomorrow s Photoshop Effects Johannes Borodajkewycz TU Wien Figure 1: Examples for two of the techniques presented in this paper: Interactive image completion with perspective correction is able to fill
More informationSingle Image Motion Deblurring Using Transparency
Single Image Motion Deblurring Using Transparency Jiaya Jia Department of Computer Science and Engineering The Chinese University of Hong Kong leojia@cse.cuhk.edu.hk Abstract One of the key problems of
More informationTexture Sensitive Image Inpainting after Object Morphing
Texture Sensitive Image Inpainting after Object Morphing Yin Chieh Liu and Yi-Leh Wu Department of Computer Science and Information Engineering National Taiwan University of Science and Technology, Taiwan
More informationBut, vision technology falls short. and so does graphics. Image Based Rendering. Ray. Constant radiance. time is fixed. 3D position 2D direction
Computer Graphics -based rendering Output Michael F. Cohen Microsoft Research Synthetic Camera Model Computer Vision Combined Output Output Model Real Scene Synthetic Camera Model Real Cameras Real Scene
More informationA Closed-Form Solution to Natural Image Matting
1 A Closed-Form Solution to Natural Image Matting Anat Levin Dani Lischinski Yair Weiss School of Computer Science and Engineering The Hebrew University of Jerusalem Abstract Interactive digital matting,
More informationDetail Preserving Shape Deformation in Image Editing
Detail Preserving Shape Deformation in Image Editing Hui Fang Google, Inc. John C. Hart University of Illinois, Urbana-Champaign (e) Figure 1: The deformation of a source image, described by tracing and
More informationLETTER Local and Nonlocal Color Line Models for Image Matting
1814 IEICE TRANS. FUNDAMENTALS, VOL.E97 A, NO.8 AUGUST 2014 LETTER Local and Nonlocal Color Line Models for Image Matting Byoung-Kwang KIM a), Meiguang JIN, Nonmembers, and Woo-Jin SONG, Member SUMMARY
More informationAutomated Removal of Partial Occlusion Blur
Automated Removal of Partial Occlusion Blur Scott McCloskey, Michael Langer, and Kaleem Siddiqi Centre for Intelligent Machines, McGill University {scott,langer,siddiqi}@cim.mcgill.ca Abstract. This paper
More informationGeometric Modeling and Processing
Geometric Modeling and Processing Tutorial of 3DIM&PVT 2011 (Hangzhou, China) May 16, 2011 6. Mesh Simplification Problems High resolution meshes becoming increasingly available 3D active scanners Computer
More informationTargil 10 : Why Mosaic? Why is this a challenge? Exposure differences Scene illumination Miss-registration Moving objects
Why Mosaic? Are you getting the whole picture? Compact Camera FOV = 5 x 35 Targil : Panoramas - Stitching and Blending Some slides from Alexei Efros 2 Slide from Brown & Lowe Why Mosaic? Are you getting
More informationManifold Preserving Edit Propagation
Manifold Preserving Edit Propagation SIGGRAPH ASIA 2012 Xiaowu Chen, Dongqing Zou, Qinping Zhao, Ping Tan Kim, Wook 2013. 11. 22 Abstract Edit propagation algorithm more robust to color blending maintain
More informationTEMPORALLY CONSISTENT REGION-BASED VIDEO EXPOSURE CORRECTION
TEMPORALLY CONSISTENT REGION-BASED VIDEO EXPOSURE CORRECTION Xuan Dong 1, Lu Yuan 2, Weixin Li 3, Alan L. Yuille 3 Tsinghua University 1, Microsoft Research Asia 2, UC Los Angeles 3 dongx10@mails.tsinghua.edu.cn,
More informationGuided Image Filtering
Guided Image Filtering Kaiming He 1, Jian Sun 2, and Xiaoou Tang 1,3 1 Department of Information Engineering, The Chinese University of Hong Kong 2 Microsoft Research Asia 3 Shenzhen Institutes of Advanced
More informationMulti-View Stereo for Static and Dynamic Scenes
Multi-View Stereo for Static and Dynamic Scenes Wolfgang Burgard Jan 6, 2010 Main references Yasutaka Furukawa and Jean Ponce, Accurate, Dense and Robust Multi-View Stereopsis, 2007 C.L. Zitnick, S.B.
More informationDynamic Color Flow: A Motion-Adaptive Color Model for Object Segmentation in Video
Dynamic Color Flow: A Motion-Adaptive Color Model for Object Segmentation in Video Xue Bai 1, Jue Wang 2, and Guillermo Sapiro 1 1 University of Minnesota, Minneapolis, MN 55455, USA 2 Adobe Systems, Seattle,
More informationDepixelizing. Pixel Art. Presented by: Jai Mashalkar ( ) Ahana Pradhan ( )
Depixelizing Pixel Art Presented by: Jai Mashalkar ( 113050007 ) Ahana Pradhan ( 113050039 ) Pixel Art http://www.pixeljoint.com/forum/forum_posts.asp?tid=11299 Use of Pixel Art Computer Games Advertising
More informationImage Segmentation Using Iterated Graph Cuts BasedonMulti-scaleSmoothing
Image Segmentation Using Iterated Graph Cuts BasedonMulti-scaleSmoothing Tomoyuki Nagahashi 1, Hironobu Fujiyoshi 1, and Takeo Kanade 2 1 Dept. of Computer Science, Chubu University. Matsumoto 1200, Kasugai,
More informationStitching and Blending
Stitching and Blending Kari Pulli VP Computational Imaging Light First project Build your own (basic) programs panorama HDR (really, exposure fusion) The key components register images so their features
More informationComputational Photography and Capture: (Re)Coloring. Gabriel Brostow & Tim Weyrich TA: Frederic Besse
Computational Photography and Capture: (Re)Coloring Gabriel Brostow & Tim Weyrich TA: Frederic Besse Week Date Topic Hours 1 12-Jan Introduction to Computational Photography and Capture 1 1 14-Jan Intro
More informationToday. Motivation. Motivation. Image gradient. Image gradient. Computational Photography
Computational Photography Matthias Zwicker University of Bern Fall 009 Today Gradient domain image manipulation Introduction Gradient cut & paste Tone mapping Color-to-gray conversion Motivation Cut &
More informationImage Composition. COS 526 Princeton University
Image Composition COS 526 Princeton University Modeled after lecture by Alexei Efros. Slides by Efros, Durand, Freeman, Hays, Fergus, Lazebnik, Agarwala, Shamir, and Perez. Image Composition Jurassic Park
More informationPanoramic Video Texture
Aseem Agarwala, Colin Zheng, Chris Pal, Maneesh Agrawala, Michael Cohen, Brian Curless, David Salesin, Richard Szeliski A paper accepted for SIGGRAPH 05 presented by 1 Outline Introduction & Motivation
More informationAn Integrated System for Digital Restoration of Prehistoric Theran Wall Paintings
An Integrated System for Digital Restoration of Prehistoric Theran Wall Paintings Nikolaos Karianakis 1 Petros Maragos 2 1 University of California, Los Angeles 2 National Technical University of Athens
More informationCapturing Skeleton-based Animation Data from a Video
Capturing Skeleton-based Animation Data from a Video Liang-Yu Shih, Bing-Yu Chen National Taiwan University E-mail: xdd@cmlab.csie.ntu.edu.tw, robin@ntu.edu.tw ABSTRACT This paper presents a semi-automatic
More informationSimultaneous Foreground, Background, and Alpha Estimation for Image Matting
Brigham Young University BYU ScholarsArchive All Faculty Publications 2010-06-01 Simultaneous Foreground, Background, and Alpha Estimation for Image Matting Bryan S. Morse morse@byu.edu Brian L. Price
More informationEfficient Edit Propagation Using Hierarchical Data Structure
JOURNAL OF L A T E X CLASS FILES, VOL. 6, NO. 1, JANUARY 2007 1 Efficient Edit Propagation Using Hierarchical Data Structure Chunxia Xiao, Yongwei Nie, Feng Tang Abstract This paper presents a novel unified
More informationFlexible Calibration of a Portable Structured Light System through Surface Plane
Vol. 34, No. 11 ACTA AUTOMATICA SINICA November, 2008 Flexible Calibration of a Portable Structured Light System through Surface Plane GAO Wei 1 WANG Liang 1 HU Zhan-Yi 1 Abstract For a portable structured
More informationAn Algorithm for Seamless Image Stitching and Its Application
An Algorithm for Seamless Image Stitching and Its Application Jing Xing, Zhenjiang Miao, and Jing Chen Institute of Information Science, Beijing JiaoTong University, Beijing 100044, P.R. China Abstract.
More informationAccelerating Double Precision FEM Simulations with GPUs
Accelerating Double Precision FEM Simulations with GPUs Dominik Göddeke 1 3 Robert Strzodka 2 Stefan Turek 1 dominik.goeddeke@math.uni-dortmund.de 1 Mathematics III: Applied Mathematics and Numerics, University
More informationStudy on Improving the Quality of Reconstructed NURBS Surfaces
Study on Improving the Quality of Reconstructed NURBS Surfaces Shufeng jiang, Shigang Wang, Yong Yan School of Mechatronic Engineering, Qiqihar University, Qiqihar 161006, China Abstract In aspect of surface
More informationFilters. Advanced and Special Topics: Filters. Filters
Filters Advanced and Special Topics: Filters Dr. Edmund Lam Department of Electrical and Electronic Engineering The University of Hong Kong ELEC4245: Digital Image Processing (Second Semester, 2016 17)
More informationSTEREO matching has been one of the most active research
> TCSVT < 1 Color Image Guided Boundary-inconsistent Region Refinement for Stereo Matching Jianbo Jiao, Student Member, IEEE, Ronggang Wang*, Member, IEEE, Wenmin Wang, Member, IEEE, Dagang Li, Member,
More informationPAPER Video Segmentation with Motion Smoothness
IEICE TRANS. INF. & SYST., VOL.E93 D, NO.4 APRIL 2010 873 PAPER Video Segmentation with Motion Smoothness Chung-Lin WEN a), Nonmember, Bing-Yu CHEN b), and Yoichi SATO c), Members SUMMARY In this paper,
More information2.1 Optimized Importance Map
3rd International Conference on Multimedia Technology(ICMT 2013) Improved Image Resizing using Seam Carving and scaling Yan Zhang 1, Jonathan Z. Sun, Jingliang Peng Abstract. Seam Carving, the popular
More informationDigital Makeup Face Generation
Digital Makeup Face Generation Wut Yee Oo Mechanical Engineering Stanford University wutyee@stanford.edu Abstract Make up applications offer photoshop tools to get users inputs in generating a make up
More informationMesh Geometric Editing Approach Based on Gpu Texture
www.ijcsi.org 67 Mesh Geometric Editing Approach Based on Gpu Texture Guiping Qian 1, YUE Wang 2 1 Assoc Prof., College of New Media, Zhejiang University of Media and Communications, Hangzhou, China 2
More informationA Feature Point Matching Based Approach for Video Objects Segmentation
A Feature Point Matching Based Approach for Video Objects Segmentation Yan Zhang, Zhong Zhou, Wei Wu State Key Laboratory of Virtual Reality Technology and Systems, Beijing, P.R. China School of Computer
More informationVideo annotation based on adaptive annular spatial partition scheme
Video annotation based on adaptive annular spatial partition scheme Guiguang Ding a), Lu Zhang, and Xiaoxu Li Key Laboratory for Information System Security, Ministry of Education, Tsinghua National Laboratory
More informationInteraction of Fluid Simulation Based on PhysX Physics Engine. Huibai Wang, Jianfei Wan, Fengquan Zhang
4th International Conference on Sensors, Measurement and Intelligent Materials (ICSMIM 2015) Interaction of Fluid Simulation Based on PhysX Physics Engine Huibai Wang, Jianfei Wan, Fengquan Zhang College
More informationStereo Matching: An Outlier Confidence Approach
Stereo Matching: An Outlier Confidence Approach Li Xu and Jiaya Jia Department of Computer Science and Engineering The Chinese University of Hong Kong {xuli,leojia}@cse.cuhk.edu.hk Abstract. One of the
More informationUse of Shape Deformation to Seamlessly Stitch Historical Document Images
Use of Shape Deformation to Seamlessly Stitch Historical Document Images Wei Liu Wei Fan Li Chen Jun Sun Satoshi Naoi In China, efforts are being made to preserve historical documents in the form of digital
More information