Mohamed Hassan

I am an AI Scientist at Electronic Arts working on learning-based character animation. I am interested in creating animatable virtual characters. I love exploring how we can create and animate them from different modalities like images, videos, text, sketches, or sparse user constraints.

I hold a Ph.D. from Max Planck Institute for Intelligent Systems, where I worked on computer vision, computer graphics and machine learning. I was advised by Michael Black. In my PhD thesis, I explored the reconstruction, analysis, and generation of Human-Scene Interaction.

During my Ph.D., I was an intern at Nvidia Toronto AI lab where I worked with Sanja Filder and Jason Peng. I also had the pleasure of doing an internship at Adobe Research in the summer of 2020. At Adobe, I worked with Duygu Ceylan, Ruben Villegas, Jun Saito, Yi Zhou, and Jimei Yang.

In my free time, I like to mentor undergraduate students from my alma mater in Sudan. I enjoy reading about history, philosophy, and economics. I also like to cycle, run, and train in martial arts.

Email  /  CV  /  Google Scholar  /  Twitter  /  Github

profile photo
  • December 2023: I gave an invited talk at the Univesity of Alberta..
  • September 2022: I joined Electronic Arts as an AI scientist working on character animation.
  • June 2021: I gave a talk about computer vision applications at the Sudanse Machine Learning Community.
  • June 2021: I joined Nvidia Toronto AI lab led by Sanja Filder for an internship.
InterPhys Synthesizing Physical Character-Scene Interactions
Mohamed Hassan, Yunrong Guo, Tingwu Wang, Michael Black, Sanja Fidler, Xue Bin Peng

Generating physical and realistic human-scene interactions without manual labeling. Generalize beyond the objects and scenarios depicted in the training dataset.

SAMP Stochastic Scene-Aware Motion Prediction
Mohamed Hassan, Duygu Ceylan, Ruben Villegas, Jun Saito, Jimei Yang, Yi Zhou, Michael Black
ICCV, 2021

Generating realistic and diverse styles of human-scene interactions.

POSA Populating 3D Scenes by Learning Human-Scene Interaction
Mohamed Hassan, Partha Ghosh, Joachim Tesch, Dimitrios Tzionas, Michael Black
CVPR, 2021

Automatically place 3D scans of people in scenes.

PSI Generating 3D People in Scenes without People
Yan Zhang, Mohamed Hassan, Michael Black, Siyu Tang
CVPR, 2019

Generates 3D human bodies in 3D scenes.

PROX Resolving 3D Human Pose Ambiguities with 3D Scene Constraints
Mohamed Hassan, Vasileios Choutas, Dimitrios Tzionas, Michael Black
ICCV, 2019

Exploit static 3D scene structure to better estimate human pose from monocular images. Introduce the PROX dataset.

Multiple proposals ArSLR Multiple proposals for continuous arabic sign language recognition
Mohamed Hassan, Khaled Assaleh, Tamer Shanableh
Sensing and Imaging, 2019

Comparison between two diferent recognition techniques for continuous Arabic Sign Language Recognition (ArSLR) for multiple users.

User-dependent ArSLR User-dependent Sign Language Recognition Using Motion Detection
Mohamed Hassan, Khaled Assaleh, Tamer Shanableh
International Conference on Computational Science and Computational Intelligence, 2016

Continuous sensor-based Arabic Sign Language Recognition (ArSLR) system based on Hidden Markov Models(HMM) and a modified version of k-nearest neighbor(KNN).

Cool website template