Pietro Perona

Bio Outline


  • Advisor
  • Learning in Machines & Brains


  • California Institute of Technology
Department of Electrical Engineering


  • United States


PhD, University of California, Berkeley
DEng, University of Padua


Pietro Perona is a neuromorphic systems engineer who studies vision: how images are processed by brains and computers in order to obtain information about the environment.

His research currently focuses on visual recognition, and how individuals detect and recognize objects and scenes in images. He is also interested in how machines could measure, represent and analyze animal and human behaviour. Perona is known for his work on partial differential equations for image processing (the Perona-Malik equation), as well as his work on unsupervised learning of visual categories and the psychophysics and modelling of human scene perception.


Longuet-Higgins Prize for Fundamental Contributions to Computer Vision, 2013

Outstanding Student Paper Award, Honorable Mention, NIPS, 2010 (student author: P. Welinder)

Koenderink Prize for Fundamental Contributions in Computer Vision, European Conference on Computer Vision, 2010

Best Paper Award, IEEE Computer Vision and Pattern Recognition Conference, 2003

NSF National Young Investigator Award, 1994

Relevant Publications

Anderson, D.J., and P. Perona. “Toward a science of computational ethology.” Neuron 84, no. 1 (2014): 18–31.

Chalupka, K., et al. “Visual Causal Feature Learning.” arXiv:1412.2309 (2014).

Perona, P. “Far and yet close: multiple viewpoints for the perfect portrait.” Art & Perception 1, no. 1–2 (2013): 105–20.

Welinder, P. et al. “The multidimensional wisdom of crowds.” Adv. Neur. In. 23 (2010): 2424–2432.

Fergus, R. et al. “Object class recognition by unsupervised scale-invariant learning.” Computer Vision and Pattern Recognition. In Proceedings of IEEE Comp. Soc., vol. 2 (June 2003).