Multidimensional signal processing for visual and multimedia services

点击翻译按钮获取中文摘要

Visual data, in the form of diverse types of imagery, remains one of the richest sources of information for both consumers and professionals. With advances in communications services and networks, and image-enabled devices, image-based services have become pervasive in recent years, especially those making use of the Internet and wireless networks. However, many of these services still offer inadequate quality, realism and responsiveness, especially for very high-resolution imagery and three-dimensional environments. The general objective of this research program is to carry out research on new multidimensional signal processing techniques that can be applied to the development of enhanced visual communications and multimedia systems. Three main themes will be pursued: multidimensional signal processing theory; color image processing and rendering; and omnidirectional and high-resolution image and video processing. The specific topics to be researched include: new more efficient multidimensional representations for different types of images and videos, including omnidirectional images; new methods for color image processing, for applications in digital cameras and color displays, among others; design of stereoscopic omnidirectional imaging systems for immersive telepresence applications; and finally, methods for video signal upconversion, such as HDTV to super HDTV conversion. In all cases, the goal is to develop fundamental new theories that lead to higher quality and/or more efficient implementation than current methods. The students trained will acquire high-value skills in image and video processing, including software development. Imaging is becoming ubiquitous in the modern world in all sectors, including business, consumer electronics and personal communications. Thus, all industries in these sectors have a strong need for personnel with skills in imaging and related fields.