Combining depth information and local edge detection for stereo image enhancement

The recent developments in 3D display technology have opened new horizons and have raised a number of challenges related to the processing and coding of 3D media. Today, stereoscopic image technology is becoming widely used in many fields. The physical limitations of image acquisition systems, however, make stereoscopic technology far from being the most widely accepted solution. Furthermore, the depth/disparity extreme ranges may subject the viewers' eyes to additional strain, causing more discomfort. To address this issue, we propose in this paper to improve stereoscopic image quality by a novel contrast enhancement method that combines local edges and depth information. The contrast is increased locally, at specific depth levels for left and right views. The increase of contrast is controlled based on the depth information, and aims at promoting the nearest objects in the 3D scene. The results obtained from a psychophysical experiment are encouraging and show that the proposed method produces stereo images that are less stressful on the eyes, thus providing more pleasant viewing experience.