ABSTRACT
Online communities and social networks are the most popular sites on the Internet, and have exploded with multimedia content in the last decade. Most web designers recognize that site images can be saved with lower fidelity to reduce bandwidth consumption and increase capacity, though many are reluctant to do so for aesthetic concerns. However, there are many images that site designers have little direct control over---socially sourced images. Many social networks automatically reduce the fidelity of uploaded images in order to conserve bandwidth. Social networks also contain a vast archive of images with popularity indicators, such as likes and shares, which recent work has correlated with psychovisual features within the images. In this paper, we investigate the trade-off between fidelity reduction and selected psychovisual enhancements. We demonstrate that even simple enhancements can be used to enable more aggressive optimization of socially sourced content, which has implications for static content delivery networks and image servers. Through user testing on real images, we validate the efficacy of our proposed approach.
- Facebook. www.facebook.com.Google Scholar
- Internet brands: vbulletin forum software. www.vbulletin.com.Google Scholar
- jquery, the write less, do more, javascript library. www.jquery.com.Google Scholar
- Curalate insights: Analysis of instagram. http://blog.curalate.com/image-qualities-that-drive-likes-on-instagram, Nov. 2013.Google Scholar
- Wikipedia: Aol art image format. http://en.wikipedia.org/wiki/ART_image_file_format, Dec. 2013.Google Scholar
- Alexa top sites. http://www.alexa.com/topsites, 2014.Google Scholar
- Curalate insights: Analysis of pinterest. http://blog.curalate.com/the-perfect-pinterest-pinhow-to-optimize-your-images-for-maximumengagement-on-pinterest, Jun. 2014.Google Scholar
- Facebook newsroom. http://newsroom.fb.com/company-info, May 2014.Google Scholar
- Facebook support: How can i make sure my photos display in the highest quality? https://www.facebook.com/help/www/266520536764594, Nov. 2014.Google Scholar
- Quintly analytics: Which facebook post type rules the news feed? http://www.quintly.com/blog/2014/03/data-analysis-which-facebook-post-type-rules-the-news-feed/, Mar. 2014.Google Scholar
- Webp: A new image format for the web. https://developers.google.com/speed/webp, March 2014.Google Scholar
- D. Beaver, S. Kumar, H. C. Li, J. Sobel, and P. Vajgel. Finding a needle in haystack: Facebook's photo storage. In USENIX OSDI, 2010. Google ScholarDigital Library
- M. Brittain. Batch processing millions of images at etsy. https://codeascraft.com/2010/07/09/batch-processing-millions-of-images, Jul. 2010.Google Scholar
- J. Cheng, L. A. Adamic, P. A. Dow, J. M. Kleinberg, and J. Leskovec. Can cascades be predicted? In 23rd International World Wide Web Conference, WWW '14, Seoul, Republic of Korea, April 7-11, 2014, pages 925--936, 2014. Google ScholarDigital Library
- S. Ding, F. Huang, Z. Xie, Y. Wu, and L. Ma. A novel customized recompression framework for massive internet images. In Proceedings of the First Int'l Conf. on Computational Visual Media, CVM'12, pages 9--16, Berlin, Heidelberg, 2012. Springer-Verlag. Google ScholarDigital Library
- eMarketer Inc. emarketer: Facebook us ad revenues. http://www.emarketer.com/Article/Facebooks-US-Ad-Revenues-Outpace-Users-Average-Daily-Time-Spent-on-Site/1011215, Sep. 2014.Google Scholar
- J. Erman, V. Gopalakrishnan, R. Jana, and K. K. Ramakrishnan. Towards a spdy'ier mobile web? In Proceedings of the Ninth ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT '13, 2013. Google ScholarDigital Library
- F. Ernawan, N. A. Abu, and N. Suryana. A psychovisual threshold for generating quantization process in tchebichef moment image compression. JCP, 9(3):702--710, 2014.Google Scholar
- H. Farid. Exposing digital forgeries from jpeg ghosts. IEEE Transactions on Information Forensics and Security, 4(1), Mar. 2009. Google ScholarDigital Library
- I. Google. Google auto enhance. http://gplusproject.appspot.com/auto-enhance/index.html, May 2013.Google Scholar
- C. Hass. Jpeg quantization tables. http://www.impulseadventure.com/photo/jpeg-quantization.html, 2008.Google Scholar
- C. Hass. Jpegsnoop - file decoding utility. http://www.impulseadventure.com/photo/jpeg-snoop.html, 2014.Google Scholar
- HTTPArchive. Tracking how the web is built. http://httparchive.org, Oct. 2014.Google Scholar
- C. Huang, A. Wang, J. Li, and K. W. Ross. Measuring and evaluating large-scale cdns paper. In Proceedings of the 8th ACM SIGCOMM Conference on Internet Measurement, IMC '08, pages 15--29. ACM, 2008. Google Scholar
- S. B. Kang, A. Kapoor, and D. Lischinski. Personalization of image enhancement. In Computer Vision and Pattern Recognition (CVPR), 2010 IEEE Conference on, pages 1799--1806, June 2010.Google ScholarCross Ref
- A. Khosla, A. D. Sarma, and R. Hamid. What makes an image popular? In 23rd International World Wide Web Conference, WWW '14, Seoul, Republic of Korea, April 7-11, 2014, pages 867--876, 2014. Google ScholarDigital Library
- M.-C. Lee and S.-B. Cho. Interactive differential evolution for image enhancement application in smart phone. In IEEE Congress on Evolutionary Computation, pages 1--6, 2012.Google Scholar
- R. K. Mantiuk, A. Tomaszewska, and R. Mantiuk. Comparison of four subjective methods for image quality assessment. Comp. Graph. Forum, 31(8), Dec. 2012. Google ScholarDigital Library
- G. Podjarny. Situational performance optimization - the next frontier. http://calendar.perfplanet.com/2012/situational-performance-optimization-the-next-frontier/, Dec. 2012.Google Scholar
- T. Shoham, D. Gill, and S. Carmel. A novel perceptual image quality measure for block based image compression. In Proceedings of SPIE, Image Quality and System Performance VIII. SPIE, 2011.Google ScholarCross Ref
- S. Stefanov. Web Performance Daybook. Oreilly and Associate Series. O'Reilly Media, Incorporated, 2012.Google Scholar
- S. Traverso, K. Huguenin, I. Trestian, V. Erramilli, N. Laoutaris, and K. Papagiannaki. Tailgate: handling long-tail content with a little help from friends. In Proceedings of the 21st World Wide Web Conference 2012, WWW 2012, Lyon, France, April 16-20, 2012, pages 151--160, 2012. Google ScholarDigital Library
- X. S. Wang, A. Balasubramanian, A. Krishnamurthy, and D. Wetherall. How speedy is spdy? In Proceedings of the 11th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2014, Seattle, WA, USA, April 2-4, 2014, pages 387--399, 2014. Google ScholarDigital Library
- M. P. Wittie, V. Pejovic, L. Deek, K. C. Almeroth, and B. Y. Zhao. Exploiting locality of interest in online social networks. In CoNEXT, page 25. ACM, 2010. Google ScholarDigital Library
Index Terms
- SERF: optimization of socially sourced images using psychovisual enhancements
Recommendations
Fractal color image compression using vector distortion measure
ICIP '95: Proceedings of the 1995 International Conference on Image Processing (Vol. 3)-Volume 3 - Volume 3A fractal monochrome image compression technique proposed by Jacquin is investigated for 24-bit true color image by encoding the RGB components' images independently. To exploit the spectral redundancy in RGB components, the root mean square error ...
Subjective evaluation of spatial resolution and quantization noise tradeoffs
Most full-reference fidelity/quality metrics compare the original image to a distorted image at the same resolution assuming a fixed viewing condition. However, in many applications, such as video streaming, due to the diversity of channel capacities ...
VSNR: A Wavelet-Based Visual Signal-to-Noise Ratio for Natural Images
This paper presents an efficient metric for quantifying the visual fidelity of natural images based on near-threshold and suprathreshold properties of human vision. The proposed metric, the visual signal-to-noise ratio (VSNR), operates via a two-stage ...
Comments