›› 2012,Vol. 27 ›› Issue (6): 1119-1128.doi: 10.1007/s11390-012-1290-4

所属专题: Artificial Intelligence and Pattern Recognition

• • 上一篇    下一篇

基于情感的灰度图彩色化

Xiao-Hui Wang1 (王晓慧), Jia Jia1 (贾珈), Han-Yu Liao2 (廖捍宇), and Lian-Hong Cai1 (蔡莲红)   

  • 收稿日期:2012-09-05 修回日期:2012-09-17 出版日期:2012-11-05 发布日期:2012-11-05

Affective Image Colorization

Xiao-Hui Wang1 (王晓慧), Jia Jia1 (贾珈), Han-Yu Liao2 (廖捍宇), and Lian-Hong Cai1 (蔡莲红)   

  1. 1. Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China;
    2. Academy of Art & Design, Tsinghua University, Beijing 100084, China
  • Received:2012-09-05 Revised:2012-09-17 Online:2012-11-05 Published:2012-11-05
  • Supported by:

    This work is supported by the National Basic Research 973 Program of China under Grant No. 2011CB302201, the National Natural Science Foundation of China under Grant Nos. 61003094, 60931160443. This work is also funded by Tsinghua National Laboratory for Information Science and Technology (TNList) Cross-Discipline Foundation of China, and supported by the Innovation Fund of Tsinghua-Tencent Joint Laboratory of China.

灰度图彩色化一直备受关注.颜色不仅能增强图像的表现力,而且是情感表达 的主要媒介.美术上常用配色方案(color theme)来表示图像的颜色构成.之前大部 分研究关注上色的结果是否符合图像语义内容,却很少考虑图像所表达的整体情感. 在自动彩色化时,同时保证语义正确以及情感符合用户需要无疑是一个挑战.在本文 中,我们提出了一套基于情感的灰度图彩色化系统.系统仅仅需要用户辅助图像分割 并给每个分割对象一个标签,以及输入一个情感词,用于指导上色的配色方案.首 先,根据标签从网上下载图像,过滤得到符合语义的参考图.然后,根据情感词,我 们基于美学原理选出一些配色方案.以这些配色方案为指导,使用遗传算法为每个对 象最终确定一幅参考图.我们提出了一种混合的纹理合成方法用于最后着色.据我们 所知,这是第一个以情感为指导,同时考虑语义的灰度图彩色化系统.同时实验证明 了系统的有效性,特别是与之前基于 MRF 的方法相比有明显优势.

Abstract: Colorization of gray-scale images has attracted many attentions for a long time. An important role of image color is the conveyer of emotions (through color themes). The colorization with an undesired color theme is less useful, even it is semantically correct. However this has been rarely considered. Automatic colorization respecting both the semantics and the emotions is undoubtedly a challenge. In this paper, we propose a complete system for affective image colorization. We only need the user to assist object segmentation along with text labels and an affective word. First, the text labels along with other object characters are jointly used to filter the internet images to give each object a set of semantically correct reference images. Second, we select a set of color themes according to the affective word based on art theories. With these themes, a generic algorithm is used to select the best reference for each object, balancing various requirements. Finally, we propose a hybrid texture synthesis approach for colorization. To the best of our knowledge, it is the first system which is able to efficiently colorize a gray-scale image semantically by an emotionally controllable fashion. Our experiments show the effectiveness of our system, especially the benefit compared with the previous Markov random field (MRF) based method.

[1] Levin A, Lischinski D, Weiss Y. Colorization using optimiza-tion. ACM Transactions on Graphics, 2004, 23(3): 689-694.

[2] Welsh T, Ashikhmin M, Mueller K. Transferring color togreyscale images. ACM Transactions on Graphics, 2002,21(3): 277-280.

[3] Ironi R, Cohen-Or D, Lischinski D. Colorization by exam-ple. In Proc. the 16th Eurographics Workshop on RenderingTechniques, June 29-July 1, 2005, pp.201-210.

[4] Charpiat G, Hofmann M, Schölkopf B. Automatic image col-orization via multimodal predictions. In Proc. the 10thECCV, October 2008, pp.126-139.

[5] Tai Y W, Jia J Y, Tang C K. Local color transfer via proba-bilistic segmentation by expectation-maximization. In Proc.CVPR2005, June 2005, Vol.1, pp.747-754.

[6] Chia A Y S, Zhuo S J, Gupta R K, Tai Y W, Cho S Y, TanP, Lin S. Semantic colorization with internet images. ACMTransactions on Graphics, 2011, 30(6), Article No.156.

[7] Arnheim R. Art and Visual Perception: A Psychology of theCreative Eye. University of California Press, 1954.

[8] Kobayashi S. Color Image Scale. Kosdansha International,1992.

[9] Kobayashi S. Art of Color Combinations. Kosdansha Inter-national, 1995.

[10] Wang X H, Jia J, Liao H Y, Cai L H. Image colorizationwith an affective word. In Proc. Computational Visual Me-dia Conference 2012, November 2012, pp.51-58.

[11] Xu K, Li Y, Ju T, Hu S M, Liu T Q. Efficient affinity-based edit propagation using K-D tree. ACM Transactionson Graphics, 2009, 28(5), Article No.118.

[12] Li Y, Ju T, Hu S M. Instant propagation of sparse edits onimages and videos. Computer Graphics Forum, 2010, 29(7):2049-2054.

[13] Huang Y C, Tung Y S, Chen J C, Wang S W, Wu J L. Anadaptive edge detection based colorization algorithm and itsapplications. In Proc. the 13th MULTIMEDIA, November2005, pp.351-354.

[14] Huang H, Zang Y, Li C F. Example-based painting guided bycolor features. The Visual Computer, 2010, 26(6/8): 933-942.

[15] Xiao X Z, Ma L Z. Gradient-preserving color transfer. Com-puter Graphics Forum, 2009, 28(6/8): 1879-1886.

[16] Reinhard E, Adhikhmin M, Gooch B, Shirley P. Color transferbetween images. IEEE Computer Graphics and Applications,2001, 21(5): 34-41.

[17] Chen T, Tan P, Ma L Q, Cheng M M, Shamir A, Hu S M.PoseShop: Human image database construction and person-alized content synthesis. IEEE Transactions on Visualizationand Computer Graphics, 2012, http://doi. ieeecomputersoci-ety.org/10.1109/TVCG.2012.148, Sept. 2012.

[18] Huang H, Zhang L, Zhang H C. Arcimboldo-like collage us-ing internet images. ACM Transactions on Graphics, 2011,30(6), Article No.155.

[19] Liu H, Zhang L, Huang H. Web-image driven best views of3D shapes. The Visual Computer, 2012, 28(3): 279-287.

[20] Chen T, Cheng M M, Tan P, Shamir A, Hu S M.Sketch2Photo: Internet image montage. ACM Transactionson Graphics, 2009, 28(5), Article No.124.

[21] Csurka G, Skaff S, Marchesotti L, Saunders C. Building look& feel concept models from color combinations: With applica-tions in image dessification, retrieval, and color transfer. TheVisual Computer, 2011, 27(12): 1039-1053.

[22] Cohen-Or D, Sorkine O, Gal R, Leyvand T, Xu Y Q. Colorharmonization. ACM Transactions on Graphics, 2006, 25(3):624-630.

[23] O'Donovan P, Agarwala A, Hertzmann A. Color compatibilityfrom large datasets. ACM Transactions on Graphics, 2011,30(4), Article No.63.

[24] Boykov Y, Funka-Lea G. Graph cuts and efficient N-D im-age segmentation. International Journal of Computer Vision,2006, 70(2): 109-131.

[25] Rother C, Kolmogorov V, Blake A. "GrabCut": Interactiveforeground extraction using iterated graph cuts. ACM Trans-actions on Graphics, 2004, 23(3): 309-314.

[26] Cheng M M, Zhang F L, Mitra N J, Huang X L, Hu S M.Repfinder: Finding approximately repeated scene elementsfor image editing. ACM Transactions on Graphics, 2010,29(4), Article, No. 83.

[27] Cheng M M, Zhang G X, Mitra N J, Huang X L, Hu SM. Global contrast based salient region detection. In Proc.CVPR2011, June 2011, pp.409-416.

[28] Belongie S, Malik J, Puzicha J. Shape matching and objectrecognition using shape contexts. IEEE Transactions on Pat-tern Analysis and Machine Intelligence, 2002, 24(4): 509-522.

[29] Wang X H, Jia J, Cai L H. Affective image adjustment witha single word. To appear in The Visual Computer.

[30] Dong Z D, Dong Q. HowNet and the Computation of Mean-ing. World Scientific, 2006.

[31] Luxburg U. A tutorial on spectral clustering. Statistics andComputing, 2007, 17(4): 395-416.

[32] Barnes C, Shechtman E, Finkelstein A, Goldman D B. Patch-match: A randomized correspondence algorithm for struc-tural image editing. ACM Transactions on Graphics, 2009,28(3), Article No.24.

[33] Chen J T, Wang B. Solid texture synthesis using position his-togram matching. In Proc. the 11th CAD/Graphics, August2009, pp.150-153.

[34] Sheikh H R, Bovik A C. Image information and visual qual-ity. IEEE Transactions on Image Processing, 2006, 15(2):430-444.

[35] Zhong F, Qin X Y, Peng Q S. Robust image segmentationagainst complex color distribution. The Visual Computer,2011, 27(6-8): 707-716.

[36] Wu J L, Shen X Y, Liu L G. Interactive two-scale color-to-gray. The Visual Computer, 2012, 28(6-8): 723-731.
No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 黎仁蔚;. Soundness and Completeness of Kung s Reasoning Procedure[J]. , 1988, 3(1): 7 -15 .
[2] Klaus Buchenrieder;. Standard-Cell Placement from Functional Descriptions[J]. , 1991, 6(1): 37 -46 .
[3] 吴信东;. Inductive Learning[J]. , 1993, 8(2): 22 -36 .
[4] 秦开怀; 范刚; 孙才;. Extrapolating Acceleration Algorithms for Finding B-Spline Intersections Using Recursive Subdivision Techniques[J]. , 1994, 9(1): 70 -85 .
[5] 王学军; 石纯一;. A Multiagent Dynamic interaction Testbed:Theoretic Framework, System Architecture and Experimentation[J]. , 1997, 12(2): 121 -132 .
[6] 董韫美;. An Interactive Learning Algorithm for Acquisition of Concepts Represented as CFL[J]. , 1998, 13(1): 1 -8 .
[7] 陆桑璐; 周笑波; 谢立;. A Model for Dynamic Adaptive Coscheduling[J]. , 1999, 14(3): 267 -275 .
[8] . L-树匹配:一种面向含噪声海量文本流的新型数据抽取模型和算法[J]. , 2005, 20(6): 763 -773 .
[9] . 采用改进型浮雕纹理映射的城区房屋模型表示方法[J]. , 2006, 21(2): 204 -208 .
[10] . 语义数据库网格的查询优化算法[J]. , 2006, 21(4): 597 -608 .
版权所有 © 《计算机科学技术学报》编辑部
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持:support@magtech.com.cn
总访问量: