This page contains projects that I participated as a researcher and as a research coordinator between 2000-2016.

CrowdRec Logo
CrowdRec: Crowd-powered recommendation for continuous digital media access and exchange in social networks

(2013-2016) Developing context-aware, interactive recommender system technology that improves recommendations by actively eliciting input from a large number of users (i.e., the crowd).

http://crowdrec.eu

sealincmedialogo
Sealinc Media: Socially-Enriched Accessed to Linked Cultural Media (COMMIT-P6)

(2012-2016) Enrichment of cultural heritage multimedia collections via linking with social media, including user-contributed images, video and annotations such as tags. (National Netherlands COMMIT project.)

http://www.commit-nl.nl/projects/socially-enriched-access-to-linked-cultural-media

CUbRIK Logo
CuBRIK: Human-enhanced time-aware multimedia search

Implementation of an open platform for multimedia search practitioners, researchers and end-users that combines the power over automatic multimedia indexing and crowdsourcing. (IP project funded by the EC)

http://www.cubrikproject.eu


Icon3
STITPROnounce: Finding Images with Verbals

Design and implementation of an open-source mobile voice application, identification of best practices for use of mobile voice interface in a low-literacy and low-technology environment. (Funded by STITPRO)

http://stitpronounce.tudelft.nl


petamedia logo

PetaMedia Network of Excellence: Peer-to-peer Tagged Media

What is the future of multimedia search technology? The PetaMedia Network of Excellence promoted the development of algorithms for multimedia retrieval and distribution based on a "Triple Synergy" combining three conventionally disparate areas of scientific investigation:
  • Multimedia content analysis
  • Analysis of user contributed tags and ratings
  • P2P relationships encoded encoded in social networks
PetaMedia brought together four national research networks in Netherlands, Switzerland, UK and Germany. (Funded by the European Commission, 2008-2011)

http://www.petamedia.eu
http://videolectures.net/peta_media/

reel of film

MultiMatch: Multilingual/Multimedia Access to Cultural Heritage

The MultiMatch project developed a search engine for the cultural heritage domain with the aim of making it possible for users to access and interact with cultural heritage content across media and across languages. An important goal of the MultiMatch project was to make the jump from presenting the user with individual documents to allowing the user to interact with complete cultural heritage entities, interconnected in context. (Funded by the European Commission, 2006-2008)

http://www.multimatch.eu


speech signal
AudioMining: Automatic methods for archival of radio broadcasts

The AudioMining project designed and implemented a prototype audio archive commissioned by Deutsche Welle and WDR, two large German radio broadcasters. The AudioMining system is MPEG-7 compliant; the necessary structural metadata are generated automatically by segmentation, classification and clustering algorithms. Vocabulary-independent audio search functionality is implemented using syllable-based speech recognition language models and a fuzzy-match search algorithm. (Industry project, 2005-2006)

Description of AudioMining project in an Ercim News Article
Current information on next generation AudioMining technology at Fraunhofer IAIS.


share project logo
SHARE: Mobile Support for Rescue Forces, Integrating Multiple Modes of Interaction

The project SHARE was an EU FP6 Information Society Technologies project developing a communication and information system that will provide emergency forces with intelligent access to multimedia information that can support the decision making process. (Funded by the European Commission, 2004-2007)


Piavida logo
Personalisierte Interaktive Audio-, Voice- und Video-Informationen fuer Portal- und Auswertungsanwendungen

The goal of π-AVIda was to develop a personalized, interactive audio/video retrieval system that integrated algorithms for the analysis, segmentation and classification of multimedia streams. (Funded by the German Federal Ministry of Education and Research, 2001-2003.)


AGMA Logo
Automatische Generierung audiovisueller Metadaten im Kontext von MPEG-7

AGMA developed a MPEG-7 compliant system for automatic generation of metadata for video and a client used for multimedia retrieval and browsing. It implemented a prototype multimedia analysis and management system for German parliamentary debates. (Funded by the German Federal Ministry of Education and Research, 2000-2003.)