My main research areas are computer vision, statistical inference, and machine learning.
I focus on developing new practical and mathematically-principled computational tools for analysis of stochastic high-dimensional real-world signals.
I'm interested in tools that provide interpretability and uncertainty quantification, that scale gracefully with the data's size, and that adapt model complexity to the data.
I'm particularly interested in Bayesian or geometric methods and problems such as unsupervised learning, motion analysis, segmentation, statistical image models, and deep learning.
Recent or current research topics (with plenty of overlap)
- Computer vision:
geometry in deep learning;
motion and tracking;
- Statistical inference and machine learning:
Bayesian methods; geometry in machine learning;
parallel and distributed computing for inference.
- Medical applications:
detection and classification of tumors (breast cancer);
epileptic seizure prediction.
Past research topics
- Applications of Lie groups and/or Riemannian manifolds
- 3D scene analysis
- Representations and statistical models of articulated pose and/or non-rigid human shape (both 2D and 3D)
- Applications of computer-vision techniques
(pose estimation, video tracking) for neuroscience
- Statistical methods for Brain MRI segmentation