Regular biography
Mathieu Salzmann is a Senior Scientist at École Polytechnique Fédérale de Lausanne (EPFL), affiliated with the CS department. His research interests include Computer Vision and Machine Learning. He is involved in teaching and PhD supervision, having directed and co-directed several EPFL theses. He also offers the course 'Introduction to Machine Learning (CS-233).' His work focuses on the principles and practical implementation of machine learning and data analysis across various scientific and application domains.
Scholar-generated biography
Mathieu Salzmann is a researcher in computer vision and machine learning, focusing on methods for understanding and modeling visual data. His work includes context-aware crowd counting, human motion prediction, and 3D human pose estimation. He has developed techniques for unsupervised domain adaptation, deep subspace clustering, and structured prediction. His research also explores kernel methods on Riemannian manifolds, compression-aware training, and geometry-aware representations. Salzmann's contributions span from learning trajectory dependencies to evaluating neural architecture search. His publications highlight the integration of deep learning with geometric and statistical modeling for robust and scalable vision systems.