A Benchmark for Image Retargeting

Michael Rubinstein	Diego Gutierrez	Olga Sorkine-Hornung	Ariel Shamir
MIT CSAIL	Universidad de Zaragoza	ETH Zurich	The Interdisciplinary Center

 

The numerous works on media retargeting call for a methodological approach for evaluating retargeting results. We present the first comprehensive perceptual study and analysis of image retargeting. First, we create a benchmark of images and conduct a large scale user study to compare a representative number of state-of-the-art retargeting methods. Second, we present analysis of the users’ responses, where we find that humans in general agree on the evaluation of the results and show that some retargeting methods are consistently more favorable than others. Third, we examine whether computational image distance metrics can predict human retargeting perception. We show that current measures used in this context are not necessarily consistent with human rankings, and demonstrate that better results can be achieved using image features that were not previously considered for this task. We also reveal specific qualities in retargeted media that are more important for viewers. The importance of our work lies in promoting better measures to assess and guide retargeting algorithms in the future. The full benchmark we collected, including all images, retargeted results, and the collected user data, are available to the research community for further investigation.

This website accompanies our paper:

A Comparative Study of Image Retargeting [pdf] [supplemental] [BibTeX] [ppt]
ACM Transactions on Graphics, Volume 29, Number 6 (Proceedings SIGGRAPH Asia 2010)

Please cite it if you use any part of the images, data, or code supplied in this website.
For questions or comments contact mrub-at-mit-at-edu

Take the survey!		Explore the dataset and results		Download		Contribute

 

News

2010-12-02	Contributor page is now open
2010-09-12	The complete dataset, results and survey system code are available for download

 

---

The authors would like to thank all who contributed to the collection and analysis of the data in this experiment. In particular, we wish to thank Yu-Shuen Wang, Yael Pritch, Eitam Kav-Venaki, Manuel Lang, Alexander Hornung and Zachi Karni for sharing their images and supplying the results of their retargeting operators, Ofir Pele for his support with the EMD metric, and Susana Castillo for her all-around help. We also wish to thank the anonymous SIGGRAPH Asia reviewers for their valuable comments. This work was supported in part by a Marie Curie grant from the Seventh Framework Programme (grant agreement no.: 251415), the Spanish Ministry of Science and Technology (TIN2010-21543), the Gobierno de Aragon (projects OTRI 2009/0411 and CTPP05/09), NYU URCF, and by gifts from TI, Microsoft Research, and Adobe Systems.