image processing

Thursday Extra 2/9: Impact of image based rendering on photography

Thursday, February 9, 2017
4:15 p.m. in Science 3821
Refreshments at 4:00 p.m. in the Computer Science Commons (Science 3817).
Everyone is welcome to attend!

Image Based Rendering and its Impact on Studio Photography is presented by Gary Meyer, Department of Computer Science and Engineering at the University of Minnesota.

A new computer graphic technique, called image based rendering, has the potential to fundamentally change how professional photographers take still life and portrait pictures. Instead of lighting the object or the person in their studio and capturing a few images, we are developing a system in which the photographer takes a comprehensive set of simple pictures with an attached flash and then uses a virtual studio lighting setup and camera to synthesize the final images. Projective texture mapping is used to blend the original pictures and project them onto a mesh that is also derived from the photographs. To achieve the desired relighting effect, the images are carefully weighted as part of the interpolation process. The result is, by definition, photorealistic because it is derived directly from the original pictures, and the method goes beyond traditional studio photography because it allows complex relighting using environment maps and it permits the production of animated flybys in addition to static shots. In addition, new lighting setups and pictures can generated any time in the future using the original set of flash photographs. The talk will be illustrated using images produced in collaboration with the Minneapolis Institute of Art.

Thursday Extra: "Robust text recognition"

On Thursday, February 10, Jerod Weinman will discuss some aspects of his recent work on text recognition:

Is your smart phone smarter than a fifth grader? Not yet. Accurately translating a photograph of text into an intrinsically textual representation has been confounding computational scientists for over a century. Humans (even fifth graders) still outperform computers at reading. In this talk, I review why the problem is difficult and present a model for robustly recognizing small amounts of text in images.

Refreshments will be served at 4:15 p.m. in the Computer Science Commons (Noyce 3817). Mr. Weinman's talk, “Robust text recognition,” will follow at 4:30 p.m. in Noyce 3821. Everyone is welcome to attend!

Syndicate content