Alvin Raj

Microgeometry capture using an elastomeric sensor

We describe a system for capturing microscopic surface geometry. The system extends the retrographic sensor [Johnson and Adelson 2009] to the microscopic domain, demonstrating spatial resolution as small as 2 microns. In contrast to existing microgeometry capture techniques, the system is not affected by the optical characteristics of the surface being measured---it captures the same geometry whether the object is matte, glossy, or transparent. In addition, the hardware design allows for a variety of form factors, including a hand-held device that can be used to capture high-resolution surface geometry in the field. We achieve these results with a combination of improved sensor materials, illumination design, and reconstruction algorithm, as compared to the original sensor of Johnson and Adelson [2009].

What your design looks like to peripheral vision

At any given instant, much of a display appears in a users periphral vision. Based on the information available in a glance, the user moves their eyes, scanning the display for items of interest, and piecing together a coherent view of the display. Much of this processing happens unconsciously. Understanding the information available in the periphery can help design better information visualizations and user interfaces, by enabling designers to make displays that effectively guide eye movements and make important information available in a glance. However, it is difficult to attend to our peripheral vision to gain insights about the information available there. In this paper, we discuss a means of visualizing the information available in peripheral vision, given an image of a display and the current fixation. We show results of our model on several information visualizations.

An interactive "retrographic sensor" for touch, texture, and shape

Retrographic sensing is a novel method for measuring surface texture and shape. It uses a sensor made of clear elastomer with a painted skin to non-destructively change an objects reflectance characteristics. When an object is pressed into the sensor, the painted skin conforms to the shape of the object. Viewed from behind, the object appears as a shaded surface and the shape of the surface can be estimated using photometric stereo techniques. In previous work [Johnson and Adelson 2009], we describe a method for reconstructing high-resolution 2.5D surface data from a single image of the sensor. In this work, we demonstrate an implementation of the system that runs at interactive rates.