The Multimedia Computing (MMC) Group conducts research on new methodological and algorithmic concepts for analyzing, interpreting, enriching, modeling, searching and recommending multimedia data available in stand-alone collections or connected in (complex) networked data structures. Learn more about the MMC Group’s research domain and expertise, impact, and research philosophy.
 

 

 

The work being carried out in the MMC Group encompasses a broad palette of research directions in the domains of multimodal analysis and processing, network data computing and multimedia information systems, with a strong influx of the expertise from the domain of human factors. Learn more about the MMC Group’s research themes.

01 June 2022

Publication award for Bishwadeep Das

Publication award for Bishwadeep Das

Biswhadeep Das received the Best Student Paper Award at the 2022 IEEE Data Science & Learning Workshop as co-writer with Elvin Isufi of the publication: "Graph Filtering Over Expanding Graphs".

26 April 2022

Symposium on Human-Centered Multimedia to celebrate Pablo Cesar’s Professorship Appointment

Symposium on Human-Centered Multimedia to celebrate Pablo Cesar’s Professorship Appointment

On May 20th 2022, Prof. Dr. Pablo Cesar will give his inaugural lecture at TU Delft, Aula, to celebrate his appointment to Professorship. The lecture, titled “Human-Centered Multimedia: making remote togetherness possible”, will provide an overview of his research with a focus on volumetric video conferencing and the future of eXtended Reality (XR) as a collaboration and communication medium.

04 April 2022

Deep learning for complex networks

Deep learning for complex networks

Image recognition, speech processing, medical diagnostics… there’s hardly an area where machine learning hasn’t made a big impact. Yet Elvin Isufi adds an extra dimension to machine learning, enabling its efficient application to complex networks such as recommender systems, social networks, and more down-to-earth water networks.

23 November 2021

Hoe de Belastingdienst lage inkomens profileerde in de jacht op fraude (Dutch)

Hoe de Belastingdienst lage inkomens profileerde in de jacht op fraude (Dutch)

Om fraude beter te kunnen opsporen begon de Belastingdienst in 2013 met een zelflerend algoritme bij de controle op toeslagen. Dat keek vooral naar de allerlaagste inkomens.

04 November 2021

Odette Scharenborg appointed vice-president of ISCA

Odette Scharenborg appointed vice-president of ISCA

Last Interspeech conference Odette Scharenborg was appointed vice-president of the International Speech Communication Association (ISCA) board, the international organisation for research on human speech recognition and speech technology.

04 November 2021

Delft researchers work on ‘inclusive automatic speech recognition’

Delft researchers work on ‘inclusive automatic speech recognition’

For the elderly, people with regional accents and non-native speakers, speech recognition systems are a pain. They don’t catch your meaning and are biased, TU research shows.

/* */