Matthias Wimmer

Papers

Total Citations

H-Index

About

Matthias Wimmer’s research lies at the intersection of computer vision, affective computing, and human-robot interaction, with a core focus on enabling machines to perceive and respond to human emotional cues. His most influential work centers on facial expression recognition, where he pioneered the use of recurrent neural networks to analyze dynamic sequences of facial movements. In his seminal 2008 paper, which has garnered 36 citations, Wimmer introduced a complete system that fits the Candide-3 face model using a learned objective function, then feeds the resulting parameter sequences into an RNN for robust classification. This approach marked a significant advance in capturing the temporal dynamics of expressions, moving beyond static image analysis. Wimmer also contributed to the practical application of this technology in human-robot interaction, as demonstrated in his 2008 prototype paper (25 citations). Furthermore, he was a key contributor to the MuDiS project, a multimodal dialogue system designed for rapid adaptation across diverse interaction scenarios, uniting expertise from computational linguistics and computer science. Wimmer’s work established foundational techniques for emotion-aware robotics, demonstrating how deep learning could bridge the gap between raw sensor data and socially intelligent machine behavior.