Currently, we work on the following projects:
  • Learned Indexing for Similarity Searching Czech Scient Foundation, lead agency project, 07/2023-06/2026 When faced with the task of storing and retrieving complex, unstructured or high-dimensional data (e.g., multimedia data), metric spaces are often employed as an underlying mathematical concept for their organization. Consequently, the only measure that can be used to arrange the data is a pairwise similarity between data objects. Similarity searching refers to a range of methods used to manage data enabling efficient search in such spaces. The main paradigm of similarity searching has remained mostly unchanged for decades — data objects are organized into a hierarchical structure according to their mutual distances, using representative pivots to reduce the number of distance computations needed to efficiently search the data. We plan to investigate an alternative to this paradigm, using machine learning models to replace pivots, thus, posing similarity search as a classification problem. We will use both supervised and unsupervised approaches to implement our solutions. We will also address the questions of scalability and dynamicity, and verify the applications for metric data.
  • VisioTherapy: Supporting physiotherapy treatments using computer-based movement analysis Technology Agency of the Czech Republic, TREND, 8/2023 – 7/2025 The goal of the project is to develop VisioTherapy software that will support patients and physiotherapists during rehabilitation therapy. The software will use AI technologies to obtain movement data from exercise videos and further compare and analyze this data. In particular, VisioTherapy will compare the patient’s current exercise with a model of the correct exercise, which will be taken in cooperation with the physiotherapist, and alert the patient to possible problems. Furthermore, VisioTherapy will monitor the progress of rehabilitation and provide summary data and statistics to both patients and physiotherapists, who will be able to use them for better targeting of treatment.
  • Searching, Mining, and Annotating Human Motion Streams Czech Science Foundation, 2019–2021 Motion capturing devices have become widely available, which resulted in large volumes of 3D human motion data produced in a variety of application domains, ranging from entertainment to medicine. However, automatized processing of such data is a challenging problem because their inherent spatio-temporal nature implies that the same action can be performed in a number of alternatives that vary in speed, timing, or location in space. Moreover, the captured data are imprecise and voluminous, as hundreds of megabytes per hour are obtained during tracking only 3D positions of body joints. Therefore, the employment of basic data-processing paradigms is much more intriguing, when compared to the traditional domains such as text or images. In the proposed project, we aim at developing new theories and technologies for three interconnected open problems of content-based searching, annotating, and mining in motion data streams. Taking into account the fast growth of motion data volumes, a lot of attention will be given to the scalability of proposed solutions.
  • CyberSecurity, CyberCrime and Critical Information Infrastructures Center of Excellence (C4e) Ministry of Education, Youth and Sports, 4/2018 – 12/2022 The project builds on the activities of the Center for Excellence in Cybercrime (C4e) and focuses on creating a new, broader and heavily integrated research strategy divided into three core research programs: Cybersecurity and Cybercrime, Critical Information Infrastructures and Secure/Safe Systems Design, and Law and Jurisprudence. Our lab contributes to the first program that consists of three related areas: (1) Simulation of advanced attacks and efficient defense, (2) Advance analysis of operational data and (3) Similarity management for big-data analytics. The first subprogram investigates methods of modelling of state-of-the-art advanced cyber attacks and their actors. Particularly, we focus on analysis of malicious code in a realistic and controlled environment and active learning of defenders. The second subprogram researches methods for aggregation and correlation of heterogeneous infrastructure events related to operation, security and performance, which enable identification of security anomalies and relational analysis of involved actors. The third subprogram aims at application of metric space paradigm of similarity for analysis of the infrastructure events. Theories, paradigms, and technologies, which make possible knowledge discovery on the large scale will be developed. All three subprograms span several cross-disciplinary topics, namely simulation, education, forensics and knowledge mining.
  • Complex Analysis and Visualization of Large-scale Heterogeneous Data Ministry of the Interior, Security Research for the Needs of the State, 2017–2020 The project aims to develop a distributed system for comprehensive analysis of big heterogeneous data – in particular, digital artifacts acquired during police investigations. Using an integrated uniform solution, the system will enable its users to analyze and visualize complex relationships and contexts of identifiable real-world entities, both by interactive data analyses, and by continuous notifications about monitored activities (based on pre-defined behavior patterns).

Selected Past Projects

  • Big Data Analytics for Unstructured Data Czech Science Foundation, 2016–2018 More information
  • Center for Large-Scale Multi-Modal Data Interpretation Czech Science Foundation, 2012–2018 More information
  • Effective Search in Large-Scale Biometric Data Ministry of Interior of the CR, 2012 – 2015 More information
  • Research of Technologies Suitable for Face Recognition That Can Be Used For Multimedia Retrieval on Internet Ministry of Education, Youth and Sports of the CR, 2012 – 2014 More information
  • Multimedia Analysis Ministry of Interior of the CR, 2010 – 2014 More information
  • Content-based Image Retrieval on the Web Scale Czech Science Foundation, 2010 – 2012 More information
  • Similarity Searching in Very Large Multimedia Databases Czech Science Foundation, 2009 – 2011 More information
  • Search on Audio-visual content using Peer-to-peer Information Retrieval (SAPIR) European Union / 6th Specific RTD Programme, 2006 – 2009 More information
  • Network of Excellence on Digital Libraries (DELOS) European Union / 6th Specific RTD Programme, 2003 – 2007 More information