Research Topics

Audio-visual Systems

  • Algorithms, components and systems for automatic audiovisual content analysis
  • Semantic image and video processing

Media Distribution and Security

  • Efficient, copyright-aware aggregation, processing and distribution of media content in B2B scenarios
  • Consideration of security issues and privacy protection in media systems

Semantic Music Technologies

  • Signal processing and pattern recognition algorithms
  • Automatic music analysis for search, recommendation, game and educational applications, and music production

Publicly Financed Research Projects

Digger - Deepfake Detection

Combining visual verification and audio forensics technologies to detect deepfakes

IMC - I-Media-Cities

Innovative e-environments for Research on Cities and the Media

ISAD

Informed Sound Activity Detection in Music Recordings

KISH

AI-based selective hearing via headphones

MICO – Media in Context

An Open-Source Framework for multi-modal Analysis

SPIRIT

Scalable privacy preserving intelligence analysis for resolving identities

Published Datasets

Within the last years the Semantic Music Technologies group has compiled audio datasets for research areas like the detection of instruments, fingerings or performance analysis. These datasets have been presented in several scientific publications on international conferences and shall serve the scientific community as potential benchmarks for comparison experiments.