The red-eye removal feature in Photoshop, which uses object recognition technology to identify eyes, was Adobe’s first foray into working with the content of images. Automatic content analysis is a major focus of computer vision and image processing research, and there is a long way to go before computers can reliably identify objects in photographs. Adobe is aiming to use the technology for image retrieval – say, finding every photo of Uncle Ralph – as well as editing and touching up pictures. The company’s goal is to give computers a two-year-old’s ability to understand the content of images, said Martin Newell, a Fellow at Adobe Systems, Inc.

Nifty research papers

A pair of imaging techniques sharpen blurry photographs. (Removing Camera Shake From a Single Photograph and Coded Exposure Photography: Motion Deblurring using Fluttered Shutter)

A smart thimble combines a force feedback sensor, accelerometer and position tracker to translate finger movements, including tracing the shapes of objects, into computer input. The fingertip digitizer can also turn any surface into a touch input device for computers, PDAs and cell phones. (Fingertip Digitizer: Applying Haptics and Biomechanics to Tactile Input Technology)

A crowd simulator for virtual environments uses the mathematics of particle flow rather than the compute-intensive current approach of attempting to model each person’s movements, perception and decisions. (Continuum Crowds)