The CNS Vision Lab conducts psychophysical and computational modeling studies of many aspects of visual perception.

A number of our projects focus on optic flow.

Some of our current and recent research projects are:

• Border-Ownership & Figure-Ground Segregation

  Border-Ownership: A model of figure-ground segregation demonstrates how cells signal border-ownership of figures in the visual scene using fast feedback connections and without junction detectors.

  Figure-Ground Segregation & Motion: A model of figure-ground segregation and motion...

• Color Perception

  What determines the color of objects that you see?

• Detecting Motion Onsets and Offsets for Kinetic Occlusion

  Surfaces can appear to have different depths even when the only information available to the visual system comes from motion.

• Heading Perception and Visually-Guided Navigation

  How do humans and other animals determine where they are going using visual information?

  Heading Perception & Independently Moving Objects: A model of visually-guided navigation explains human heading bias in the presence of independently moving objects.

  Motion Processing: Moving through the environment provides many useful cues. Relative depth is provided by motion parallax. The direction of ego-motion can be estimated from optical flow. Kinetic boundaries allow for the segregation of the scene into static and moving objects. Our aim is to build a model that combines all these cues to define the steering control for visual navigation.

  The ViSTARS Model: A model of motion processing neurons demonstrates how primates, including humans, determine their direction of travel from visual input. The model links neurophysiology with psychophysics and technology.

• Hyperacuity with complex stimuli

  Simple hyperacuity phenomena can be modeled with a small number of oriented filters. Such models break down when the stimuli become more complex, such as opposite polarity with respect to the background, or masked by sinusoidal gratings, or when the gap between stimulus elements is increased. How can these stimuli be modeled?

• Strategic and Fluid Allocation of Visual Attention

  Given a limited amount of attention to distribute to the objects that we see, what is the best way to allocate attention to optimize performance?

• Stereoscopic Viewing for Medical X-Ray Images

  New stereoscopic 3-D viewing software using graphics card computing to improve methods for search in radiology. This software will improve speed and accuracy for detection of dangerous lung nodules.

• The Watercolor Effect

  Modeling The Watercolor Effect.