Università della Svizzera italiana Home

Computer vision

Computer vision is an image science that uses methods for acquiring, processing, analyzing, and understanding images and, in general, high-dimensional data from the real world in order to produce numerical or symbolic information. Methods employed depend on the application and rely on vision analysis functions such as image acquisition, feature extraction, mesh processing, detection, segmentation, pattern recognition, decision making, all elements that are contained within the normal human visual sense.  In fact, computer vision attempts to recreate and enhance human vision for applications such as navigation, modeling, control processes, and decision-making.

Research topics include:

  • Geometry: theoretical and computational methods in metric geometry, embedding problems, discrete Gromov-Hausdorff distances, spectral and diffusion geometry, geometric approaches to problems in image sciences.
  • Machine learning: manifold learning, non-linear dimensionality reduction, multidimensional scaling, metric learning, similarity-sensitive hashing, multimodal data fusion, multimodal metric learning.
  • Computer graphics and shape analysis: non-rigid similarity and correspondence, partial similarity, symmetry, feature based methods, large-scale shape retrieval, invariant texture mapping, shape synthesis and morphing.
  • Computer vision: 3D acquisition and reconstruction, feature descriptors, image and video retrieval and search, copy detection, use of bioinformatics algorithms for video analysis, Internet-scale applications.
  • Biometrics: 3D face recognition.

Link to Knowledge Area: Faculty of Informatics

Knowledge Area Leader: Michael Bronstein


    All events