Advanced personalized web applications require a carefully dealing with their users ’ wishes and preferences. Since such preferences do not always hold in general, personalized applications also have to consider the user’s current situation. In this paper we present a novel framework for modeling situations and situated preferences. Our approach consists of a general meta model for situations, which can be applied as foundation for situation models in a wide range of applications. Furthermore, an XML-based preference repository for the storage and management of situated preferences is developed. Long-term and situated preferences can easily be accessed with the preference repository interface. Particularly, preferences best-matching to a given situation can be queried. This approach allows web applications to react flexibly and personalized to the changing situations of their users.

Our experiments in TRECVID 2005 include participation in the high-level feature extraction and search tasks. In the highlevel feature extraction task, we applied a method of representing semantic concepts as class models on a set of parallel Self-Organizing Maps (SOMs). We submitted one run, A PicSOM 1, in which we applied a feature selection scheme for each concept separately. The results showed that the SOM-based class models can be used for representing semantic concepts on multimodal feature indices and that the proposed method is suitable for detecting video shots with specific semantic content. In the search task, we submitted a total of seven runs (three automatic, three manual, and one interactive run). Our main motivation was to study the utilization of parallel multimodal features and class models compared to using only text-based queries. The overall settings for the runs were as follows: • F A 1 SOM-F1 7: a baseline automatic run using only ASR/MT output • F A 2 SOM-F2 3: an automatic run using ASR/MT output, multimodal features, and class models • F A 2 SOM-F3 5: an automatic run using multimodal features and class models • M A 1 SOM-M1 6: a baseline manual run using only ASR/MT output • M A 2 SOM-M2 4: a manual run using ASR/MT output and multimodal features • M A 2 SOM-M3 2: a manual run using ASR/MT output, multimodal features, and class models • I A 2 SOM-I 1: an interactive run Both in the automatic and manual experiments, we observed that the proposed method is able to combine the text query, multimodal features and class models successfully. In both cases, the overall best results are obtained using all three information sources with the MAP value being nearly double when compared to text-only search. Our small-scale interactive search experiments were performed with our prototype retrieval interface supporting only relevance feedback-based retrieval. Still, the experiments demonstrate that the proposed method can also be used in an interactive setting, where the search is guided with iterative feedback from the user.

Similarity search in metric spaces is a general paradigm that can be used in several application fields. It can also be effectively exploited in content-based image retrieval systems, which are shifting their target towards the Web-scale dimension. In this context, an important issue becomes the design of scalable solutions, which combine parallel and distributed architectures with caching at several levels. To this end, we investigate the design of a similarity cache that works in metric spaces. It is able to answer with exact and approximate results: even when an exact match is not present in cache, our cache may return an approximate result set with quality guarantees. By conducting tests on a collection of one million high-quality digital photos, we show that the proposed caching techniques can have a significant impact on performance, like caching on text queries has been proved effective for traditional Web search engines.

KDDI R&D Laboratories has been participating in the past TREC conferences for text retrieval tasks. In this year we are newly participating Video Retrieval Evaluation at TREC 2003, shot boundary determination task and story segmentation task. The report on our story segmentation task will be presented on site. In shot boundary determination task, we applied our proprietary shot segmentation algorithm originally proposed in [1] and slightly

In this paper we examine how Self-Organizing Maps (SOMs) can be used in detecting and describing emergent ontological relations between semantic objects and object classes in a visual database. The ontological relations we have studied include co-existence, taxonomies of visual and semantic similarity and spatial relationships. The used database contains 2618 images, each of which belongs to one or more of ten predefined semantic classes. 1

the input data is a collection of images that are annotated with a given keyword, such as “car”. The problem is to attribute the annotation to specific parts of the images. There exists plenty of suitable input data readily

Abstract. We present in this paper an integrated system that allows the management and annotation of multimedia objects stored in MPEG-7/21 repositories, and the specification and semi-automatic generation of multimedia presentations based on the content relationships existing among multimedia objects. This system is the outcome of the collaboration

In this thesis, the problems associated with the encoding and transcoding of multiple video objects are considered. In contrast to frame-based video processing, the handling of multiple video objects introduces new degrees of freedom that can be exploited in terms of compression capabilities and transmission through the network. Additionally, new sources of information must be considered, such as the boundary definition of an object.

Content-based image retrieval (CBIR) systems can be used also for other purposes than online access to unannotated image databases. In particular, when a CBIR system is equipped with an automatic image segmentation subsystem, keyword annotations given on image level can be focused on specific image segments. In this paper, we show that our PicSOM CBIR system is able to reveal semantic knowledge not only from keyword annotations but also from recorded online use of the system. This automatically extracted high abstraction level visual information can then be used to further improve the accuracy of the system and to categorize the objects of the database with semantic concepts. This process, we claim, then helps to bridge the semantic gap between low-level visual features available for computers and the high-level semantic terms used by the humans. The results of the experiments described in this paper support that view. 1.

In this paper we present a prototype for parental control that detects images with adult content received on a mobile device. More specifically, the application that we developed is able to intercept images received through various communication channels (bluetooth, MMS) on mobile devices based on the Symbian T M operating systems. Once intercepted, the images are analysed by the component of the system that automatically classify images with explicit sexual content. At the current stage the application that intercept images runs on the mobile device, the classifier runs on a remote server. computational capabilities, mobile phones provide little or no protection against undesired contents exchange. This aspect becomes crucial when the user of the mobile phone is a minor. While the inspection of transferred contents is currently possible for messages moving through the infrastructured part of the network, the analysis of locally exchanged messages is still infeasible. Short range, point-to-point connections, for instance using bluetooth or IR links, create small, temporary networks providing no guarantees on the sent data.