›› 2012, Vol. 27 ›› Issue (6): 1129-1139.doi: 10.1007/s11390-012-1291-3

Special Issue: Artificial Intelligence and Pattern Recognition

• Special Section on Computational Visual Media • Previous Articles     Next Articles

A Customized Framework to Recompress Massive Internet Images

Shou-Hong Ding1 (丁守鸿), Fei-Yue Huang2 (黄飞跃), Zhi-Feng Xie1 (谢志峰), Yong-Jian Wu2 (吴永坚), Bin Sheng1 (盛斌), and Li-Zhuang Ma1,* (马利庄), Member, CCF   

  1. 1. Department of Computer Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China;
    2. Tencent Research, Shanghai 200233, China
  • Received:2012-09-05 Revised:2012-10-06 Online:2012-11-05 Published:2012-11-05
  • Supported by:

    This work is supported by a joint project of Tencent Research and Shanghai Jiao Tong University. It is also partially supported by the National Basic Research 973 Program of China under Grant No. 2011CB302203, the National Natural Science Foundation of China under Grant Nos. 61073089, 61133009, the Open Projects Program of National Laboratory of Pattern Recognition of China, and the Open Project Program of the State Key Lab of CAD&CG of Zhejiang University of China under Grant No. A1206.

Recently, device storage capacity and transmission bandwidth requirements are facing a heavy burden on account of massive internet images. Generally, to improve user experience and save costs as much as possible, a lot of internet applications always focus on how to achieve appropriate image recompression. In this paper, we propose a novel framework to efficiently customize image recompression according to a variety of applications. First of all, we evaluate the input image's compression level and predict an initial compression level which is very close to the final output of our system using a prior learnt from massive images. Then, we iteratively recompress the input image to different levels and measure the perceptual similarity between the input image and the new result by a block-based coding quality method. According to the output of the quality assessment method, we can update the target compression level, or switch to the subjective evaluation, or return the final recompression result in our system pipeline control. We organize subjective evaluations based on different applications and obtain corresponding assessment report. At last, based on the assessment report, we set up a series of appropriate parameters for customizing image recompression. Moreover, our new framework has been successfully applied to many commercial applications, such as web portals, e-commerce, online game, and so on.

[1] Chen T, Cheng M M, Tan P, Shamir A, Hu S M.Sketch2Photo: Internet image montage. ACM Trans. Graph-ics, 2009, 28(5), Article No.124.

[2] Huang H, Zhang L, Zhang H C. Arcimboldolike collage usinginternet images. ACM Trans. Graphics, 2011, 30(6), ArticleNo.155.

[3] Zhuang Y, Han Y, Wu F, Yang J. Stable multi-label boostingfor image annotation with structural feature selection. Sci-ence China Information Sciences, 2011, 54(12): 2508-2521.

[4] Yang F, Li B. Unsupervised learning of spatial structuresshared among images. The Visual Computer, 2011, 27(2):175-180.

[5] Xie Z F, Lau R, Gui Y, Chen M G, Ma L Z. A gradient-domain-based edge-preserving sharpen filter. The VisualComputer, Published Online: 19 Jan., 2012.

[6] Pennebaker W B, Mitchell J L. JPEG: Still Image Data Com-pression Standards. Springer, 1992.

[7] Rabbani M, Joshi R. An overview of the JPEG 2000 still im-age compression standard. Signal Processing: Image Com-munication, 2002, 17(1): 3-48.

[8] Taubman D. High performance scalable image compressionwith EBCOT. IEEE Trans. Image Processing, 2000, 9(7):1158-1170.

[9] Do M, Vetterli M. The contourlet transform: An efficient di-rectional multiresolution image representation. IEEE Trans-actions on Image Processing, 2005, 14(12): 2091-2106.

[10] He X, Ji M, Bao H. A unified active and semi-supervised learn-ing framework for image compression. In Proc. Int. Conf.Computer Vision and Pattern Recognition, June 2009, pp.65-72.

[11] Ierodiaconou S, Byrne J, Bull D, Redmill D, Hill P. Unsu-pervised image compression using graphcut texture synthesis.In Proc. the 16th IEEE International Conference on ImageProcessing, November 2009, pp.2289-2292.

[12] Byrne J, Ierodiaconou S, Bull D, Redmill D, Hill P. Unsuper-vised image compression-by-synthesis within a JPEG frame-work. In Proc. the 15th IEEE International Conference onImage Processing, October 2008, pp.2892-2895.

[13] Wang Z, Bovik A, Sheikh H, Simoncelli E. Image quality as-sessment: From error visibility to structural similarity. IEEETransactions on Image Processing, 2004, 13(4): 600-612.

[14] Chandler D, Hemami S. VSNR: A waveletbased visual signal-to-noise ratio for natural images. IEEE Transactions on Im-age Processing, 2007, 16(9): 2284-2298.

[15] Shnayderman A, Gusev A, Eskicioglu A. An SVD-basedgrayscale image quality measure for local and global assess-ment. IEEE Trans. Image Processing, 2006, 15(2): 422-429.

[16] Shoham T, Gill D, Carmel S. A novel perceptual image qual-ity measure for block based image compression. In Proc.SPIE 7867, Farnand S, Gaykema F (eds.), January 2011,pp.786 709-786 715.

[17] Hamberg R, de Ridder H. Continuous assessment of time-varying image quality. In Proc. SPIE 3016, Rogowitz B,Pappas T (eds.), June 1997, pp.248-259.

[18] de Ridder H. Psychophysical evaluation of image quality:From judgment to impression. In Proc. SPIE 3299, RogowitzB E, Pappas T N (eds.), July 1998, pp.252-263.

[19] Sheikh H, Sabir M, Bovik A. A statistical evaluation of re-cent full reference image quality assessment algorithms. IEEETransactions on Image Processing, 2006, 15(11): 3440-3451.

[20] International Telecommunication Union — Radiocommunica-tion Sector. Studies towards the unification of picture assess-ment methodologies. Technical Report, BT.1082-1, 1990.

[21] International Telecommunication Union — Radiocommunica-tion Sector. Methodology for the subjective assessment of thequality of television pictures. Technical Report, BT.500-11Recommendation, 2003.

[22] Ding S, Huang F, Xie Z,Wu Y, Ma L. A novel customizedrecompression framework for massive internet images. InProc. Computational Visual Media Conference, November2012, pp.9-16.

[23] Liu D, Sun X, Wu F, Li S, Zhang Y Q. Image compressionwith edge-based inpainting. IEEE Transactions on Circuitsand Systems for Video Technology, 2007, 17(10): 1273-1287.

[24] Cheng L, Vishwanathan S V N. Learning to compress imagesand videos. In Proc.the 24th International Conference onMachine Learning, June 2007, pp.161-168.

[25] Sheikh H R, Wang Z, Cormack L, Bovik A C. Qual-ity assessment database. Release 2, 2005, http://live.ece.utexas.edu/research/quality.

[26] Liu Y J, Luo X, Xuan Y M et al. Image retargeting qualityassessment. Computer Graphics Forum, 2011, 30(2): 583-592.

[27] Wang Z, Bovik A. Mean squared error: Love it or leave it? anew look at signal fidelity measures. IEEE Signal ProcessingMagazine, 2009, 26(1): 98-117.

[28] Sampat M, Wang Z, Gupta S, Bovik A, Markey M. Complexwavelet structural similarity: A new image similarity index.IEEE Trans. Image Processing, 2009, 18(11): 2385-2401.

[29] Wang Z, Simoncelli E. Translation insensitive image similar-ity in complex wavelet domain. In Proc. IEEE Interna-tional Conference on Acoustics, Speech, and Signal Process-ing, March 2005, pp.573-576.

[30] Wang Z, Li Q. Information content weighting for perceptualimage quality assessment. IEEE Transactions on Image Pro-cessing, 2011, 20(5): 1185-1198.

[31] Sheikh H, Bovik A. Image information and visual quality.IEEE Trans. Image Processing, 2006, 15(2): 430-444.

[32] Bauschke H, Hamilton C, Macklem M, McMichael J, SwartN. Recompression of JPEG images by requantization. IEEETransactions on Image Processing, 2003, 12(7): 843-849.

[33] Ng C, Ng V, Poon P. Quantisation error reduction for re-ducing Q-factor JPEG recompression. In Proc. IFSA WorldCongress and 20th NAFIPS International Conference, July2001, pp.1460-1465.

[34] International Telecommunication Union —Radiocommunica-tion Sector. Subjective assessment methods for image qualityin high definition television. Technical Report, BT.710-4 Rec-ommendation, 1998.
No related articles found!
Full text



[1] Liu Mingye; Hong Enyu;. Some Covering Problems and Their Solutions in Automatic Logic Synthesis Systems[J]. , 1986, 1(2): 83 -92 .
[2] Chen Shihua;. On the Structure of (Weak) Inverses of an (Weakly) Invertible Finite Automaton[J]. , 1986, 1(3): 92 -100 .
[3] Gao Qingshi; Zhang Xiang; Yang Shufan; Chen Shuqing;. Vector Computer 757[J]. , 1986, 1(3): 1 -14 .
[4] Chen Zhaoxiong; Gao Qingshi;. A Substitution Based Model for the Implementation of PROLOG——The Design and Implementation of LPROLOG[J]. , 1986, 1(4): 17 -26 .
[5] Huang Heyan;. A Parallel Implementation Model of HPARLOG[J]. , 1986, 1(4): 27 -38 .
[6] Min Yinghua; Han Zhide;. A Built-in Test Pattern Generator[J]. , 1986, 1(4): 62 -74 .
[7] Tang Tonggao; Zhao Zhaokeng;. Stack Method in Program Semantics[J]. , 1987, 2(1): 51 -63 .
[8] Min Yinghua;. Easy Test Generation PLAs[J]. , 1987, 2(1): 72 -80 .
[9] Zhu Hong;. Some Mathematical Properties of the Functional Programming Language FP[J]. , 1987, 2(3): 202 -216 .
[10] Li Minghui;. CAD System of Microprogrammed Digital Systems[J]. , 1987, 2(3): 226 -235 .

ISSN 1000-9000(Print)

CN 11-2296/TP

Editorial Board
Author Guidelines
Journal of Computer Science and Technology
Institute of Computing Technology, Chinese Academy of Sciences
P.O. Box 2704, Beijing 100190 P.R. China
E-mail: jcst@ict.ac.cn
  Copyright ©2015 JCST, All Rights Reserved