Advances in Region-of-Interest Video and Image Processing

Abstract

The advent of cheaper and more powerful devices with the ability to play, create, and transmit video content has led to a dramatic increase in the multimedia content distribution on both wireline and wireless networks. Also, the reduction of cost of digital video cameras along with the development of user-generated video sites (e.g., iTunes™, YouTube™) stimulated a new user-generated video content sector and made unprecedented demands for high-quality and low-delay video communication. The Region-of-Interest (ROI) is a desirable feature in many future scalable video coding applications, such as mobile device applications, which have to be adapted to be displayed on a relatively small screen; thus, a mobile device user may wish to extract and track only a predefined ROI within the displayed video. At the same time, other users having a larger mobile device screen may wish to extract other ROIs to receive higher video stream resolution. Therefore, to fulfill these requirements, it would be beneficial to simultaneously transmit or store a video stream in a variety of ROIs, as well to enable efficiently tracking of the predefined Region-of-Interest. This chapter presents recent advances in Region-of-Interest video and image processing techniques for multimedia applications, while making a special emphasis on a scalable extension of the H.264/AVC standard. The detailed observations and conclusions, which are presented in this chapter, are supported by authors’ personal experience in this field, thereby presenting a variety of experimental results.

Analysis‐aware microscopy video compression

“…The video compression method described in this article and in our earlier work can be characterized as region‐of‐interest (ROI) based methods. Previous ROI video encoding methods have been explored in Grois and Hadar (), Liu, Li, and Soh (), Van Leuven, Van Schevensteen, Dams, and Schelkens (). One application of ROI video coding to face detection and tracking is discussed in Menser and Brunig ().…”

Analysis and implementation of an artificial vision method for micropositioning based on a camera-screen system

“…Es necesario crear una máscara de una región de interés (ROI) para enfocar el procesado en una región particular de una imagen. Los avances en las técnicas de procesamiento de imágenes y vídeo de ROI has sido estudiados por diferentes autores como Grois [139], Shah [140] o Zhou [141]. Una máscara de imagen es una imagen que contiene valores de 1 y 0.…”