ion --- VIR systems differ in the level of abstraction in which content is indexed. For example, images may be indexed at various levels, such as at the feature-level (e.g., color, texture, and shape), object-level (e.g., moving foreground object), syntax-level (e.g., video shot), and semantic-level (e.g., image subject), and so forth. Most automatic VIR systems aim at lowlevel features, while the high-level indexes are usually generated manually. Interaction among different levels is an exciting but unsolved issue. . Generality --- VIR systems differ in their specificity of the domain of visual information. For example, customized feature sets can be developed to incorporate specific domain knowledge, such as those in medical and remote-sensing applications. Other, more general VIR systems aim at indexing unconstrained visual information such as that on the Internet. . Content Collection --- VIR systems differ in the methods in which new visual information is ad...

Previous research in conceptual indexing methods of images has furnished us with refined theoretical frameworks characterising various aspects of images that could and should be indexed using textual descriptors. The development of digital image processing technologies has bred a brigade of content-based indexing and retrieval methods available for applications. What the users need and in what kinds of environments different indexing and retrieval methods are relevant, has remained an area of less intensive research work.

This paper provides an overview of current research in image information retrieval and provides an outline of areas for future research. The approach is broad and interdisciplinary and focuses on three aspects of image research (IR): text-based retrieval, content-based retrieval, and user interactions with image information retrieval systems. The review concludes with a call for image retrieval evaluation studies similar to TREC. Keywords: Information Science, Image Retrieval, CBIR, Introduction Interest in image retrieval has increased in large part due to the rapid growth of the World Wide Web. According to a recent study (Lawrence & Giles, 1999) there are 180 million images on the publicly indexable Web, a total amount of image data of about 3Tb [terabytes], and an astounding one million or more digital images are being produced every day (Jain, 93). The need to find a desired image from a collection is shared by many groups, including journalists, engineers, historians, designers...

This paper is an introduction to the track framework -- the tasks, data, and measures -- and the approaches taken. An overview of results will be presented in the plenary session. For detailed information about approaches and results see the various site reports and the back-of-the-notebook results pages for the Video Track

In order to resolve the mismatch between user needs and current image retrieval techniques, we conducted a study to get more information about what users look for in images. First, we developed a framework for the classification of image descriptions by users, based on various classification methods from the literature. The classification framework distinguishes three related viewpoints on images, namely nonvisual metadata, perceptual descriptions and conceptual descriptions. For every viewpoint a set of descriptive classes and relations is specified. We used the framework in an empirical study, in which image descriptions were formulated by 30 participants. The resulting descriptions were split into fragments and categorized in the framework. The results suggest that users prefer general descriptions as opposed to specific or abstract descriptions. Frequently used categories were objects, events and relations between objects in the image.

Content creation, editing, and searching are extremely time consuming tasks that often require substantial training and experience, especially when high-quality audio and video are involved. "New media" represents a new paradigm for multimedia information representation and processing, in which the emphasis is placed on the actual content. It thus brings the tasks of content creation and searching much closer to actual users and enables them to be active producers of audiovisual information rather than passive recipients. We discuss the state-of-the-art and present next-generation techniques for content representation, searching, creation, and editing. We discuss our experiences in developing a Web-based distributed compressed video editing and searching system (WebClip), a media representation language (Flavor) and an object-based video authoring system (Zest) based on it, and large image/video search engines for the World-Wide Web (WebSEEk and VideoQ). We also present a case study of new media applications based on specific planned multimedia education experiments with the above systems in several K-12 schools in Manhattan.

This paper reports the results of a field study on journalists' practices in requesting, searching for and selecting photos in the course of their daily work. The study addresses different types of search topics common in journalistic illustration tasks, journalists' searching behaviour and the criteria they apply in selecting photos. Data were collected by observing journalists in their work and interviewing them. A sample of requests received by the archive was also analysed. The results indicate that specific needs dominate the use of newspaper photo archives. Photos of objects, themes, or abstract topics expressed in general terms were also needed, but finding them and formulating queries in these cases especially was considered problematic. The results suggest that browsing is an essential strategy in accessing digital photo archives.

TRECVID 2005 represented the fifth running of a TREC-style video retrieval evaluation, the goal of which remained to promote progress in content-based retrieval from digital video via open, metrics-based

Abstract. This paper investigates the applicability of high-level semantic features for video retrieval using the benchmarked data from TRECVID 2003 and 2004, addressing the contributions of features like outdoor, face, and animal in retrieval, and if users can correctly decide on which features to apply for a given need. Pooled truth data gives evidence that some topics would benefit from features. A study with 12 subjects found that people often disagree on the relevance of a feature to a particular topic, including disagreement within the 8 % of positive feature-topic associations strongly supported by truth data. When subjects concur, their judgments are correct, and for those 51 topicfeature pairings identified as significant we conduct an investigation into the best interactive search submissions showing that for 29 pairs, topic performance would have improved had users had access to ideal classifiers for those features. The benefits derive from generic features applied to generic topics (27 pairs), and in one case a specific feature applied to a specific topic. Re-ranking submitted shots based on features shows promise for automatic search runs, but not for interactive runs where a person already took care to rank shots well. 1

The TREC Video Retrieval Evaluation (TRECVID) is an annual international e#ort, funded by the US Advanced Research and Development Agency (ARDA) and the National Institute of Standards and Technology (NIST) to promote progress in contentbased retrieval from digital video via open, metricsbased evaluation. Now beginning its fourth year, TRECVID aims over time to develop both a better understanding of how systems can e#ectively accomplish video retrieval and how one can reliably benchmark their performance. This paper is a case study in the development of video retrieval systems and their evaluation as well as a report on the TRECVID status to-date. After an introduction to the evolution of TRECVID over the past 3 years, we report on the most recent evaluation TRECVID 2003 in terms of the 4 tasks (shot boundary determination, highlevel feature extraction, story segmentation and classification, search), the data (133 hours of US television news), the measures, the results obtained, and the approaches taken by some of the 24 participating groups.