Biomedical Computer Vision at UniAndes

Our field of research is computer vision, the area of artificial intelligence seeking automated understanding of visual information, and the interactions between its three fundamental problems, or Malik's 3Rs of vision: perceptual organization (or re-organization), recognition and reconstruction. Perceptual organization refers to the ability of the human visual system to group together local image entities that belong together, e.g., to the same texture pattern, the same contour, or the same object. Recognition implies connecting visual data with our prior knowledge of the world, and its tasks include classifying images and pixels into semantic visual categories. Reconstruction consists in recovering geometric structure from multiple views or different imaging modalities of the same object or scene. The interactions between these three problems constitute the current frontier of research in computer vision. 

Our main domain of application is quantitative biomedical image analysis. We aim at leveraging recent progress in computer vision to contribute to the solution of high-impact and large-scale biological and medical problems of the real world. Our ongoing projects include creating a 3D virtual atlas of Drosophila embryogenesis, low-cost automated tuberculosis diagnosis with portable microscopy, and detection and reconstruction of proteins with electron microscopy.



pablo arbelaez2

Pablo Arbelaez

Associate Professor

Department of Biomedical Engineering

pa.arbelaez 'at'

Office ML-430, Ext. 1783

Google Scholar

Semantic Scholar







  • 08/2017: 6 Published articles (Oral Presentations) at SIPAIM.




  • 01/2017: Published article in TPAMI.


  • 11/2016: Postdoc position in biomedical computer vision available. Contact Pablo Arbelaez for further details.


  • 09/2016: Ph.D. opportunities in biomedical computer vision for top students. Contact Pablo Arbelaez for further details.