News & Events

12-14-16

Pietro Perona, Allen E. Puckett Professor of Electrical Engineering, and colleagues have developed the Merlin Bird Photo ID mobile app which uses machine-learning technology to identify hundreds of North American bird species it "sees" in photos. "This app is the culmination of seven years of our students' hard work and is propelled by the tremendous progress that computer-vision and machine-learning scientists are making around the world," says Professor Perona. "A machine that recognizes objects in images, like humans do, was a distant dream when I was a graduate student and now it's finally happening." [Caltech story]

Tags: EE research highlights CMS Pietro Perona

07-27-16

Professor Pietro Perona, has developed a method using Google Earth and Google Street View to count the trees in the city of Los Angeles. The process of counting the trees using human tree counters is very expensive and would cost about $3 million today. The last time the city did such counting was more than two decades ago and at the time there were 700,000 street trees. Perona has tested the methodology in a section of Pasadena where the city recently commissioned a sidewalk survey. By comparing the results to the known inventory, he determined that the computer was about 80% accurate. [LA Times story] [KPCC story]

Tags: EE research highlights CMS Pietro Perona

06-11-15

Carlos Roberto Gonzalez, advised by Professors Abu-Mostafa and Perona is the winner of this year’s Charles Wilts Prize, for his doctoral thesis "Optimal Data Distributions in Machine Learning." The Charles Wilts Prize is awarded every year to a graduate student in Electrical Engineering for outstanding independent research.

Tags: EE honors Pietro Perona Wilts Prize Yaser Abu-Mostafa Carlos Palacios

04-15-14

Professor Babak Hassibi’s students Kishore Jaganathan, and Christos Thrampoulidis as well as Professor Pietro Perona’s students Ron Appel, and Krzysztof Chalupka, have won the 2014 Qualcomm Innovation Fellowship. Jaganathan, and Thrampoulidis’ proposal is entitled Interference Alignment via Matrix Completion for Cellular Networks and Network Coding. Appel, and Chalupka’s proposal is entitled Energy-Efficient Multiclass Classification for Visual Applications on Mobile Devices.  Each winner will receive a $100K fellowship. This year there were 137 submissions and only 9 winners have been announced. Caltech is the only school to have two winning teams. [List of Winners]

Tags: EE honors Pietro Perona Babak Hassibi Kishore Jaganathan Christos Thrampoulidis Ron Appel Krzysztof Chalupka

09-30-13

Pietro Perona, Allen E. Puckett Professor of Electrical Engineering, and colleague's paper entitled "Object Class Recognition by Unsupervised Scale-Invariant Learning" has received the Longuet-Higgins Prize of the IEEE Computer Society. The prize is given at the Conference on Computer Vision and Pattern Recognition (CVPR), for fundamental contributions in Computer Vision. The award recognizes CVPR papers from ten years ago with fundamental impact on computer vision research. [List of Past Recipients]

Tags: EE Pietro Perona

09-27-12

Pietro Perona, Allen E. Puckett Professor of Electrical Engineering, and colleagues have shown that the distance at which facial photos are taken influences perception. Their study found that close-up photo subjects are judged to look less trustworthy, less competent, and less attractive. [Caltech Release]

Tags: EE research highlights Pietro Perona

04-13-12

Pietro Perona, Allen E. Puckett Professor of Electrical Engineering, and colleagues including graduate student, Peter Welinder, have been selected for the Innovation Corps (i-Corps) program of the National Science Foundation (NSF). The aim of the i-Corps program, which was highlighted by the NSF Director at a recent Wouk Lecture, is to guide promising research with commercial potential out of university laboratories. The winning Caltech proposal is entitled "Combining Machine Vision and Crowdsourcing for Convenient and Accurate Image Annotation." The team has proposed to combine the complementary strengths of human annotators and machines into a hybrid system that would annotate a large body of images which would be a valuable in scientific, medical, as well as many commercial applications. [Video of Wouk Lecture]

Tags: EE research highlights health Pietro Perona NSF Peter Welinder

04-02-10

Christof Koch, Lois and Victor Troendle Professor of Cognitive and Behavioral Biology and Professor of Computation and Neural Systems, Pietro Perona, Allen E. Puckett Professor of Electrical Engineering, and colleagues' research on decision making and visual saliency has found that human decisions are influenced by both value and saliency in a way that is consistent with the ideal Bayesian observer. [Abstract]

Tags: EE research highlights Pietro Perona Christof Koch

12-02-09

The collaboration of Pietro Perona, Allen E. Puckett Professor of Electrical Engineering, Michael Dickinson Esther M. and Abe M. Zarem Professor of Bioengineering, and David J. Anderson, Seymour Benzer Professor of Biology, is highlighted in a Nature article entitled "Flies on film". [Nature Article]

Tags: EE Michael Dickinson Pietro Perona

04-08-09

Researchers led by Pietro Perona, the Allen E. Puckett Professor of Electrical Engineering, and David J. Anderson, the Roger W. Sperry Professor of Biology and a Howard Hughes Medical Institute Investigator, have trained computers to automatically analyze aggression and courtship in fruit flies, opening the way for researchers to perform large-scale, high-throughput screens for genes that control these innate behaviors. The program allows computers to examine half an hour of video footage of pairs of interacting flies in what is almost real time; characterizing the behavior of a new line of flies "by hand" might take a biologist more than 100 hours. "This is a coming-of-age moment in this field," says Perona. "By choosing among existing machine vision techniques, we were able to put together a system that is much more capable than anything that had been demonstrated before." This work is detailed in the April issue of Nature Methods. [Caltech Press Release]

Tags: EE research highlights Pietro Perona