Hayley Hung
Hayley Hung researches automatic understanding of human behavior and social interaction. Her career includes a 2007 doctorate at Queen Mary, meeting-analysis research at Idiap and a fellowship on distant social-interaction analysis in Amsterdam. At Delft, her work combines cameras, microphones and wearable sensors. She developed the MINGLE research direction and helped release MatchNMingle in 2018. With collaborators, she later tested neural methods for identifying conversational groups and roles from wearable measurements. The work studies observable behavior.
- Works
- 148
- Citations
- 3,341
- h-index
- 27
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
2007 onward
From video behavior to group interaction
Hung completed her doctorate, From Visual Saliency to Video Behaviour Understanding, in 2007. She then researched dominance and influence in meetings at Idiap, followed by an Amsterdam fellowship on social interactions viewed from a distance.
-
2016–2017
The MINGLE research project
In May 2016, MINGLE received a Vidi grant and was planned to start in 2017. The project concerns social group dynamics and interaction quality in complex scenes, using multiple sensors and nonverbal behavior.
-
2018
Release MatchNMingle
Hung and collaborators released MatchNMingle in June 2018. The dataset supports study of interactions during real-life mingling events, providing a research resource for social perception and multimodal behavior analysis.
-
2019
Detect conversations from wearables
Hung, Alessio Rosatelli and Ekin Gedik studied identifying conversational group membership and speaker/listener roles from wearable sensors. Their model combines proximity and acceleration and was tested on MatchNMingle, showing the value of combining the two signals in that setting.
Sources and credits
Photo: ELLIS profile; creator not stated. License: Not stated by source. Source: ellis.eu. Cropped to a square and resized. Ask us to remove or replace this photo.