Jan Sedmidubsky

A Self-Organized System for Content-Based Search in Multimedia

We propose a self-organized and self-adapting system for content-based search in multimedia data. In particular, we build a semantic overlay over an existing peer-to-peer network. The self-organization of the overlay is obtained by using the social-network paradigm. The connections between peers are formed on the basis of a query-answer principle. The knowledge about answers to previous queries is exploited to route queries efficiently. At the same time, a randomized mechanism is used to explore new and unvisited parts of the network. In this way, the self-adaptable and robust system is built. Moreover, the metric space data model is used to achieve extensibility. The proposed concepts are verified on a network consisting of 2,000 peers and indexing 10 million images.

Adaptive Approximate Similarity Searching through Metric Social Networks

Exploiting the concepts of social networking represents a novel approach to the approximate similarity query processing. We present a metric social network where relations between peers, giving similar results, are established on per-query basis. Based on the universal law of generalization, a new query forwarding algorithm is proposed. The same principle is used to manage query histories of individual peers with the possibility to tune the tradeoff between the extent of the history and the level of the query-answer approximation. All algorithms are tested on real data and real network of computers.

A Key-Pose Similarity Algorithm for Motion Data Retrieval

Analysis of human motion data is an important task in many research fields such as sports, medicine, security, and computer animation. In order to fully exploit motion databases for further processing, effective and efficient retrieval methods are needed. However, such task is difficult primarily due to complex spatio-temporal variances of individual human motions and the rapidly increasing volume of motion data. In this paper, we propose a universal content-based subsequence retrieval algorithm for indexing and searching motion data. The algorithm is able to examine database motions and locate all their sub-motions that are similar to a query motion example. We illustrate the algorithm usability by indexing motion features in form of joint-angle rotations extracted from a real-life 68-minute human motion database. We analyse the algorithm time complexity and evaluate retrieval effectiveness by comparing the search results against user-defined ground truth. The algorithm is also incorporated in an online web application facilitating query definition and visualization of search results.

MUFIN: A Multi-feature Indexing Network

It has become customary that practically any information can be in a digital form. However, searching for relevant information can be complicated because of: (1) the diversity of ways in which specific data can be sorted, compared, related, or classified, and (2) the exponentially increasing amount of digital data. Accordingly, a successful search engine should address problems of extensibility and scalability. The Multi-Feature Indexing Network (MUFIN) is a general purpose search engine that satisfies these requirements. The extensibility is ensured by adopting the metric space to model the similarity, so MUFIN can evaluate queries over a wide variety of data domains compared by metric distance functions. The scalability is achieved by utilizing the paradigm of structured peer-to-peer networks, where the computational workload of query execution is distributed over multiple independent peers which can work in parallel. We demonstrate these unique capabilities of MUFIN on a database of 100 million images indexed according to a combination of five MPEG-7 descriptors.

Gait Recognition Based on Normalized Walk Cycles

We focus on recognizing persons according to the way they walk. Our approach considers a human movement as a set of trajectories formed by specific anatomical landmarks, such as hips, feet, shoulders, or hands. The trajectories are used for the extraction of distance-time dependency signals that express how a distance between a pair of specific landmarks on the human body changes in time as the person walks. The collection of such signals characterizes a gait pattern of person’s walk. To determine the similarity of gait patterns, we propose several functions that compare various combinations of extracted signals. The gait patterns are compared on the level of individual walk cycles in order to increase the recognition effectiveness. The results evaluated on a 3D database of walking humans achieved the recognition rate up to 96 %.

Assessing similarity models for human-motion retrieval applications

The development of motion capturing devices poses new challenges in the exploitation of human‐motion data for various application fields, such as computer animation, visual surveillance, sports, or physical medicine. Recently, a number of approaches dealing with motion data have been proposed, suggesting characteristic motion features to be extracted and compared on the basis of similarity. Unfortunately, almost each approach defines its own set of motion features and comparison methods; thus, it is hard to fairly decide which similarity model is the most suitable for a given kind of human‐motion retrieval application. To cope with this problem, we propose the human motion model evaluator, which is a generic framework for assessing candidate similarity models with respect to the purpose of the target application. The application purpose is specified by a user in form of a representative sample of categorized motion data. Respecting such categorization, the similarity models are assessed from the effectiveness and efficiency points of view using a set of space‐complexity, information‐retrieval, and performance measures. The usability of the framework is demonstrated by case studies of three practical examples of retrieval applications focusing on recognition of actions, detection of similar events, and identification of subjects. Copyright

Query Routing Mechanisms in Self-Organizing Search Systems

We analyze routing mechanisms of a self-organizing semantic overlay for content-based search in multimedia data. This overlay operates over any existing P2P network based on the metric space approach. In particular, we replace the previous design of routing mechanisms in Metric Semantic Overlay (MSO) with a new adaptive query-routing algorithm. An advantage of it lies in an automatic tuning of confusability of queries that is used to select peers during query evaluation. These improvements are experimentally evaluated on a real-life and synthetic dataset.

Identifying walk cycles for human recognition

We concentrate on recognizing persons according to the way they walk. Our approach considers a human movement as a set of trajectories of hips, knees, and feet captured as the person walks. The trajectories are used for the extraction of viewpoint invariant planar signals that express how a distance between a pair of specific points on the human body changes in time. We solely focus on analysis and normalization of extracted signals to simplify their similarity comparison, without presenting any specific gait recognition method. In particular, we propose a novel method for automatic determination of walk cycles within extracted signals and evaluate its importance on a real-life human motion database.

Similarity Searching: Towards Bulk-Loading Peer-to-Peer Networks

Due to the exponential growth of digital data and its complexity, we need a technique which allows us to search such collections efficiently. A suitable solution seems to be based on the peer-to-peer (P2P) network paradigm and the metric-space model of similarity. During the building phase of the distributed structure, the peers often split as new peers join the network. During a peer split, the local data is halved and one half is migrated to the new peer. In this paper, we study the problem of efficient splits of metric data locally organized by an M-tree and we propose a novel algorithm for speeding the splits up. In particular, we focus on the metric-based structured P2P network called the M-Chord. In experimental evaluation, we compare the proposed algorithm with several straightforward solutions on a real network organizing 10 million images. Our algorithm provides a significant performance boost.

Querying similarity in metric social networks

In this paper we tackle the issues of exploiting the concepts of social networking in processing similarity queries in the environment of a P2P network. The processed similarity queries are laying the base on which the relationships among peers are created. Consequently, the communities encompassing similar data emerge in the network. The architecture of the presented metric social network is formally defined using the acquaintance and friendship relations. Two version of the navigation algorithm are presented and thoroughly experimentally evaluated. Finally, learning ability of the metric social network is presented and discussed.