Scene flow estimation methodologies and applications — A review
Xiang Xuezhi,Syed Masroor Ali,Mingliang Zhai,Deguang Xiao
DOI: https://doi.org/10.1109/CCDC.2017.7979461
2017-01-01
Abstract:Image processing is an inevitable tool for visual tracking. Visual object tracking is a very hot area of research in the computer vision. Computer vision tasks include methods for acquiring, processing, analyzing and understanding digital images, and in general, deal with the extraction of high-dimensional data from the real world in order to produce numerical or symbolic information, e.g., in the forms of decisions. In this core area, Scene Flow(3-D) Estimation is one of the most significant algorithm used in numerous applications like it can be used as image processing techniques, controlling processes, e.g., an industrial robot, as navigation, e.g., by an autonomous vehicle or mobile robot, as detecting events, e.g., for visual surveillance or people counting, as organizing information, e.g., for indexing databases of images and image sequences, as modeling objects or environments, e.g., medical image analysis or topographical modeling, as interaction, e.g., as the input to a device for computer-human interaction, and automatic inspection, e.g., in manufacturing applications. This indicates that the Scene flow (3-D) Estimation algorithm hold potential for more sophisticated and controlled applications in fields of Visual Tracking/Computer Vision. The main theme of this paper is to provide knowledge regarding the latest techniques of most known applications of Scene Flow/Scene Flow Estimations. Extensive references are provided for more in depth explanation.