Many users of image retrieval systems would prefer to express initial queries using keywords. However, manual keyword indexing is very time-consuming. Therefore, a content-based image retrieval system which can automatically assign keywords to images would be very attractive. Unfortunately, it has proved very challenging to build such systems, except where either the image domain is restricted or the keywords relate only to low-level concepts such as color. This article presents a novel image indexing and classification system, called CLAIRE (CLAssifying Images for REtrieval), composed of one image processing module and three modules of support vector machines for color, texture, and high-level concept classification for keyword assignment. The experimental prototype system described here assigns up to five keywords selected from a controlled vocabulary of 60 terms to each image. The system is trained offline by 1639 examples from the Corel stock photo library. For evaluation, five judges reviewed a sample of 800 unknown images to identify which automatically assigned keywords were actually relevant to the image. The system proved to have an 80% probability to assign at least one relevant keyword to an image.Full Paper: http://doi.acm.org/10.1145/1165774.1165777 Full Proceedings: ACM Transactions on Information Systems (TOIS), Volume 24, Issue 3 (July 2006).
If you also have not been to ASSETS 2007, the proceedings are now available at the ACM DL and the best papers are:
Best paper: Evaluating American Sign Language generation through the participation of native ASL signers
We discuss important factors in the design of evaluation studies for systems that generate animations of American Sign Language (ASL) sentences. In particular, we outline how some cultural and linguistic characteristics of members of the American Deaf community must be taken into account so as to ensure the accuracy of evaluations involving these users. Finally, we describe our implementation and user-based evaluation (by native ASL signers) of a prototype ASL generator to produce sentences containing classifier predicates, frequent and complex spatial phenomena that previous ASL generators have not produced.Best student paper: Slipping and drifting: using older users to uncover pen-based target acquisition difficulties
This paper presents the results of a study to gather information on the underlying causes of pen -based target acquisition difficulty. In order to observe both simple and complex interaction, two tasks (menu and Fitts' tapping) were used. Thirty-six participants across three age groups (18-54, 54-69, and 70-85) were included to draw out both general shortcomings of targeting, and those difficulties unique to older users. Three primary sources of target acquisition difficulty were identified: slipping off the target, drifting unexpectedly from one menu to the next, and missing a menu selection by selecting the top edge of the item below. Based on these difficulties, we then evolved several designs for improving pen-based target acquisition. An additional finding was that including older users as participants allowed us to uncover pen-interaction deficiencies that we would likely have missed otherwise.Full Proceedings: Proceedings of the 9th international ACM SIGACCESS conference on Computers and accessibility
Labels: assets2007, image, Information retrieval