Esam Ghaleb

EsamProfile2022_edt.JPG

Since September 2024, I am a researcher at the Multimodal Language Department at the Max Planck Institute for Psycholinguistics in Nijmegen, the Netherlands. In this position, I will continue my collaboration with Prof. Asli Ozyurek on multimodal communication and behavior analysis.

Research Background & Interest. I am an AI researcher with a background in computer science. My research is driven by my fascination with how people use verbal and nonverbal behaviors in their interactions, from expressing and perceiving emotions to building shared understanding. In my research, I combine insights from cognitive science, psycholinguistics and psychology to computationally model and understand the complexity of human interaction. During my Ph.D. and postdoctoral work at Maastricht University, I made methodological and technical contributions to multimodal modeling of individual behaviors, such as emotions and activities, notably extending explainable AI for emotion recognition to include gestural expressions.Before joining the Multimodal Language Department at the Max Planck Institute for Psycholinguistics, I was a postdoctoral researcher at the Institute for Logic, Language & Computation,  University of Amsterdam. There, I focused on modeling and understanding linguistic and gestural alignment in face-to-face dialogues and the automatic segmentation and representation of co-speech gestures.

Beside, I have built a strong track record in applied research with societal impact, particularly at the interface of AI and healthcare and e-learning. In these areas, I worked on two EU projects and co-organized pilot studies with over 200 users. Additionally, I led a work package in an EU project that focused on fusing the socio-health expertise of healthcare professionals with machine intelligence. My overarching goal is to integrate verbal and nonverbal behaviors into interactive technologies that prioritize human needs, creating inclusive communication platforms that advance assistive technologies for people with diverse abilities.

Academic Background. I am a reseacher with international profile. Born in Yemen, I graduated in the top 0.2% of high school graduates nationwide, securing a competitive full scholarship to pursue my bachelor’s and master’s degrees in computer engineering at Istanbul Technical University, where I graduated with honors in both programs. I further enriched my international profile during my master’s program with an internship at the Karlsruhe Institute of Technology in Germany. Already as a junior researcher, I published an open-access dataset on age-invariant face recognition using Harry Potter movies. I then pursued my Ph.D. in computer science at Maastricht University, where I developed novel computational methodologies for multimodal emotion recognition.

News

Oct 2, 2024 I am overjoyed to share the news of the birth of our son, Elias Ghaleb, on October 2, 2024. Both mother and baby are healthy and doing well.
Aug 16, 2024 Paper accepted at the 26th ACM International Conference on Multimodal Interaction!
May 29, 2024 Research Talk @ Maastrich University on Visual Modeling of Human Behaviors: Multimodal, Interpretable, Grounded, and Self-Supervised Approaches
May 27, 2024 Guest Lecture on Body Language Modeling in the Master of AI, NLP2 Course at the UvA
May 24, 2024 Seminar Presentation: Co-Speech Gesture Modeling at Tilburg University

Selected Publications

  1. Learning Co-Speech Gesture Representations in Dialogue through Contrastive Learning: An Intrinsic Evaluation
    Esam Ghaleb, Bulat Khaertdinov, Wim Pouw, and 4 more authors
    In International Conference on Multimodal Interaction 2024
  2. Leveraging Speech for Gesture Detection in Multimodal Communication
    Esam Ghaleb, Ilya Burenko, Marlou Rasenberg, and 7 more authors
    arXiv preprint arXiv:2404.14952 2024
  3. Speakers align both their gestures and words not only to establish but also to maintain reference to create shared labels for novel objects in interaction
    Sho Akamine, Esam Ghaleb, Marlou Rasenberg, and 3 more authors
    In Proceedings of the Annual Meeting of the Cognitive Science Society 2024
  4. Analysing Cross-Speaker Convergence in Face-to-Face Dialogue through the Lens of Automatically Detected Shared Linguistic Constructions
    Esam Ghaleb, Marlou Rasenberg, Wim Pouw, and 4 more authors
    In Proceedings of the Annual Meeting of the Cognitive Science Society 2024
  5. Co-Speech Gesture Detection through Multi-phase Sequence Labeling (to appear)
    Esam Ghaleb, Ilya Burenko, Marlou Rasenberg, and 6 more authors
    In IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2024
  6. Joint Modelling of Audio-visual Cues Using Attention Mechanism for Emotion Recognition
    Esam Ghaleb, Jan Niehues, and Stylianos Asteriadis
    Multimedia Tools and Applications 2023
  7. Skeleton-Based Explainable Bodily Expressed Emotion Recognition Through Graph Convolutional Networks
    Esam Ghaleb, André Mertens, Stylianos Asteriadis, and 1 more author
    In 2021 16th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2021) 2021