Personalized photograph ranking and selection system
Authors:
Open Access
- 25 October 2010
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 211-220
Abstract
In this paper, we propose a novel personalized ranking system for amateur photographs. Although some of the features used in our system are similar to previous work, new features, such as texture, RGB color, portrait (through face detection), and black-and-white, are included for individual preferences. Our goal of automatically ranking photographs is not intended for award-wining professional photographs but for photographs taken by amateurs, especially when individual preference is taken into account. The performance of our system in terms of precision-recall diagram and binary classification accuracy (93%) is close to the best results to date for both overall system and individual features. Two personalized ranking user interfaces are provided: one is feature-based and the other is example-based. Although both interfaces are effective in providing personalized preferences, our user study showed that example-based was preferred by twice as many people as feature-based.Keywords
This publication has 22 references indexed in Scilit:
- Photo and Video Quality Evaluation: Focusing on the Subject2008
- Color harmonizationACM Transactions on Graphics, 2006
- Studying Aesthetics in Photographic Images Using a Computational Approach2006
- Efficient Graph-Based Image SegmentationInternational Journal of Computer Vision, 2004
- Image quality assessment: from error visibility to structural similarityIEEE Transactions on Image Processing, 2004
- Classification of Digital Photos Taken by Photographers or Home Users2004
- MPEG-7 Homogeneous Texture DescriptorETRI Journal, 2001
- Support vector learning for ordinal regression1999
- Texture features for browsing and retrieval of image dataIEEE Transactions on Pattern Analysis and Machine Intelligence, 1996
- Ordinal Measures of AssociationJournal of the American Statistical Association, 1958