A fundamental goal of vision science is to understand visual function under natural conditions. However, natural stimuli are extremely complicated and difficult to characterize mathematically. As a result, most vision research has investigated visual processing with artificial stimuli- e.g. bars and blobs- that are easier to characterize. This approach has led to huge gains in our understanding of human and animal vision systems, but it has also left important gaps in our knowledge.

The primary aim of my research is to enable the principled, rigorous study of critical visual tasks under natural conditions. To accomplish this aim, I want to understand i) the computations that optimally transform sensory information into behaviorally relevant representations of the environment and ii) the computations that humans and animals actually use.

I develop tools to enable rigorous mathematical characterization of natural. I apply these tools to the study of specific natural tasks. We are currently obtaining results that can be used to generate principled, testable hypotheses about the processing of visual information on the level of neural circuits. Many times, the methods that are developed for the study of one task in one organism can be applied to similar tasks in other organisms. Sometimes, the methods can also be profitably applied to technological problems. For example, we are applying our current work on defocus blur estimation to the problem of auto-focusing digital cameras.

My current work is focused on understanding how to estimate depth from natural images. I have developed computational methods for optimally estimating individual depth cues in natural images (e.g. defocus, disparity, motion). I use psychophysical methods to determine how accurately humans estimate those same depth cues. I am also planning computational and psychophysical studies that will investigate the combination of those depth cues in natural images. Previous work has included research on natural scene statistics, wave-optics, cue-combination, cue-calibration, and visuo-motor adaptation.

Name the animals! Answers are upside down: