Roland Angst @ Stanford University >> Research >> Research Interests

Research Interests

My primary area of expertise is geometric computer vision. However, in order to achieve my research goals, I am exploring a broad set of topics, amongst others:

  • Computer Vision
    • Multiple-view geometry and structure-from-motion
    • Subspace-based analysis and inference (aka. factorization methods)
    • 3D scene understanding
  • Machine learning
    • Sampling techniques (sequential monte carlo, markov chain monte carlo, RANSAC, ...)
    • Non-parametric Bayesian models
    • Missing data and outliers (e.g. in collaborative filtering)
    • Sparsity-inducing models (L1-norm, trace-norm, Dirichlet-processes, ...)
  • Mathematical interests
    • Linear and multilinear algebra
    • Matrix and tensor factorizations and decompositions
    • Optimization techniques (e.g. convex and global optimization, ...)
    • Compressive sensing

As the Major of my Master studies was Computer Graphics, I also like to keep an eye on certain topics in graphics, in particular on computational photography and physically based simulations in computer graphics.

© Roland Angst