My research is in computational vision, in particular how to combine bottom-up processing, such as image segmentation with top-down information, such as recognition of familiar shapes. I'm interested in how measuring the predictive power of different visual cues can provide general information-theoretic constraints on human visual processing. I also work on developing tools for biological image analysis in order to measure morphology and spatial patterns of gene expression in developing animals.

news:

• The second SoCal Computer Vision Meetup will be held at UCI October 30th.
  
  
• H. Pirsiavash, D. Ramanan, C. Fowlkes. "Bilinear classifiers for visual recognition", NIPS (Dec. 2009).

• C. Desai, D. Ramanan, C. Fowlkes. "Discriminative models for multi-class object layout", ICCV Kyoto, Japan, (Sept. 2009). [pdf]     *Awarded the Marr Prize for best paper*

• P. Arbelaez, M. Maire, C. Fowlkes and J. Malik. "From Contours to Regions: An Empirical Evaluation" CVPR Miami Beach, FL, (June 2009). [pdf]
  Latest segmentation and boundary detection results here with code here.
  Fast GPU implementation and web service here.
  
  
• C. Fowlkes, et. al. "A Quantitative Spatiotemporal Atlas of Gene Expression in the Drosophila Blastoderm," Cell, 133(2), 364-374.   (reviews of this paper also appeared in Developmental Cell and Nature Methods)
  
  
• The initial release of our Drosophila blastoderm 3D gene expression atlas is online. Take a look around or download the visualization tool and give it a whirl.

teaching:

cs 177 : applications of probability in computer science [spring 08]
cs 216 : image understanding [fall 08]
cs 117 : project in computer vision [winter 08]
cs 295 : research topics in vision [spring 09]