A5072233507- Video Analysis and Summarization
- Advanced Image and Video Retrieval Techniques
- Image Retrieval and Classification Techniques
- Multimodal Machine Learning Applications
- Music and Audio Processing
- Semantic Web and Ontologies
- Multimedia Communication and Technology
- Advanced Vision and Imaging
- Natural Language Processing Techniques
- Digital Rights Management and Security
- Domain Adaptation and Few-Shot Learning
- Video Surveillance and Tracking Methods
- Advanced Neural Network Applications
- Generative Adversarial Networks and Image Synthesis
- Library Science and Information Systems
- Face recognition and analysis
- Human Pose and Action Recognition
- Digital and Traditional Archives Management
- Video Coding and Compression Technologies
- Web Data Mining and Analysis
- Image and Video Quality Assessment
- Robotics and Sensor-Based Localization
- Time Series Analysis and Forecasting
- Digital Humanities and Scholarship
- Digital Media Forensic Detection
Joanneum Research
2016-2025
Istituto di Scienza e Tecnologie dell'Informazione "Alessandro Faedo"
2023
HTW Berlin - University of Applied Sciences
2023
Charles University
2023
University of Klagenfurt
2023
University of Graz
2010
For the seventh time since 2018, Lifelog Search Challenge (LSC) benchmarked interactive lifelog search systems in a live challenge. The LSC goal is to comparatively evaluate system capabilities access large multimodal lifelogs comprising hundreds of thousands records. LSC'24 attracted an unprecedented record number twenty-one participating teams, where each team proposes innovative ideas implemented new or already established retrieval systems. benchmark was organised front audience at...
The last decade has seen innovations that make video recording, manipulation, storage, and sharing easier than ever before, thus impacting many areas of life. New retrieval scenarios emerged as well, which challenge the state-of-the-art approaches. Despite recent advances in content analysis, can still benefit from involving human user loop. We present our experience with a class interactive methodology to stimulate evolution new More specifically, browser showdown evaluation campaign is...
Despite the fact that automatic content analysis has made remarkable progress over last decade - mainly due to significant advances in machine learning interactive video retrieval is still a very challenging problem, with an increasing relevance practical applications. The Video Browser Showdown (VBS) annual evaluation competition pushes limits of state-of-the-art tools, tasks, data, and metrics. In this paper, we analyse results outcome 8th iteration VBS detail. We first give overview novel...
For the sixth time since 2018, Lifelog Search Challenge (LSC) was organized as a comparative benchmarking exercise for various interactive lifelog search systems. The goal of this international competition is to test system capabilities access large multimodal lifelogs. LSC'23 attracted twelve participanting teams, each whom had developed competitive retrieval system. benchmark in front live audience at LSC workshop ACM ICMR'23. As previous editions, introductory paper presents and...
This paper conducts a thorough examination of the 12th Video Browser Showdown (VBS) competition, which is well-established international benchmarking campaign for interactive video search systems. The annual VBS competition has witnessed steep rise in popularity multimodal embedding-based approaches retrieval. majority thirteen systems participating 2023 utilized CLIP-based cross-modal model, allowing specification free-form text queries to visual content. shared emphasis on joint embedding...
Interactive video retrieval tools developed over the past few years are emerging as powerful alternatives to automatic approaches by giving user more control well responsibilities. Current research tries identify best combinations of image, audio and text features that combined with innovative UI design maximize performance. We present last installment Video Browser Showdown 2015 which was held in conjunction International Conference on MultiMedia Modeling (MMM 2015) has stated aim pushing...
This work summarizes the findings of 7th iteration Video Browser Showdown (VBS) competition organized as a workshop at 24th International Conference on Multimedia Modeling in Bangkok. The focuses video retrieval scenarios which searched scenes were either previously observed or described by another person (i.e., an example shot is not available). During event, nine teams competed with their tools providing access to shared collection 600 hours content. Evaluation objectives, rules, scoring,...
Comprehensive and fair performance evaluation of information retrieval systems represents an essential task for the current age. Whereas Cranfield-based evaluations with benchmark datasets support development models, significant efforts are required also user-oriented that try to boost interactive search approach. This article presents findings from 9th Video Browser Showdown, a competition focuses on legitimate comparison designed challenging known-item tasks over large video collection....
Neural Network Coding and Representation (NNR) is the first international standard for efficient compression of neural networks (NNs). The designed as a toolbox methods, which can be used to create coding pipelines. It either an independent framework (with its own bitstream format) or together with external network formats frameworks. For providing highest degree flexibility, methods operate per parameter tensor in order always ensure proper decoding, even if no structure information...
MPEG-7 is an excellent choice for the description of audiovisual content due to its flexibility and comprehensiveness. The drawback that these properties also increase complexity descriptions cause ambiguities which hinder interoperability. In order partly solve problems, profiles levels have been proposed, but definitions adopted lack semantic constraints are necessary We propose a profile detailed can be used in broad range applications. aims at describing single multimedia entities,...
For enabling immersive user experiences for interactive TV services and automating camera view selection framing, knowledge of the location persons in a scene is essential. We describe an architecture detecting tracking high-resolution panoramic video streams, obtained from Omni Cam, stitching streams 6 HD resolution tiles. use CUDA accelerated feature point tracker, blob detector HOG person detector, which are used region each tiles before fusing results entire panorama. In this paper we...
The digitization initiatives in the past decades have led to a tremendous increase digitized objects cultural heritage domain. Although digitally available, these are often not easily accessible for interested users because of distributed allocation content different repositories and variety data structure standards. When search content, they first need identify specific repository then know how within this platform (e.g., usage vocabulary). goal EEXCESS project is design implement an...
Diminished reality (DR) refers to the removal of real objects from environment by virtually replacing them with their background. Modern DR frameworks use inpainting hallucinate unobserved regions. While recent deep learning-based is promising, case complicated need generate coherent structure and 3D geometry (i.e., depth), in particular for advanced applications, such as scene editing. In this paper, we propose Deep DR, a first RGB-D framework fulfilling all requirements DR: Plausible image...
The Video Browser Showdown (VBS) is a live video browsing competition where international researchers, working in the field of interactive search, evaluate and demonstrate efficiency their tools presence audience. aim VBS to for at known-item search (KIS) tasks with well-defined data set direct comparison other tools.
The Video Browser Showdown (VBS) has influenced the Multimedia community already for 10 years now. More than 30 unique teams from over 21 countries participated in VBS since 2012 already. In 2021, we are celebrating 10th anniversary of VBS, where 17 international compete against each other an unprecedented contest fast and accurate multimedia retrieval. this tutorial discuss motivation details contest, including its history, rules, evaluation metrics, achievements We talk about properties...
We report preliminary results of PrestoPRIME, an EU FP7 integrated project, including audiovisual (AV) archives, academics and industrial partners, focused on long-term digital preservation AV media objects improving access by integrating archives with European on-line libraries, specifically Europeana. Project outcomes will result in tools services to ensure the permanence content museums other collections, enabling future dynamically changing contexts. PrestoPRIME has a special focus...