3-D Camera Movements Tracking and Indexing Based on 2-D Hints Extraction

Abstract

Objects and camera movements are important clues which can be used for a better video and image understanding for content-based image and video indexing. In this paper, we propose a new general technique for 3-D camera movement tracking based on 2-D hints extracted from 2-D images. Our approach consists of extracting first 2-D hints based on object contours and calculating their different order derivatives. We then apply our pattern matching method to obtain objects movement vectors, and with help of 3-D projection theory, we detect the camera movement description in the 3-D space. The experiments that we have conducted on two video sequences show that our technique characterizes accurately the 3-D camera movement. The further work we have been undertaking shows that 2-D and 3-D hints combined with movement vectors can lead to 3-D scene description.