3D Motion Analysis via Energy Minimization

This work deals with 3D motion analysis from stereo image sequences for driver assistance systems. It consists of two parts: the estimation of motion from the image data and the segmentation of moving objects in the input images. The content can be summarized with the technical term machine visual kinesthesia, the sensation or perception and cognition of motion. In the first three chapters, the importance of motion information is discussed for driver assistance systems, for machine vision in general, and for the estimation of ego motion. The next two chapters delineate on motion perception, analyzing the apparent movement of pixels in image sequences for both a monocular and binocular camera setup. Then, the obtained motion information is used to segment moving objects in the input video. Thus, one can clearly identify the thread from analyzing the input images to describing the input images by means of stationary and moving objects. Finally, I present possibilities for future applications based on the contents of this thesis. Previous work in each case is presented in the respective chapters. Although the overarching issue of motion estimation from image sequences is related to practice, there is nothing as practical as a good theory (Kurt Lewin). Several problems in computer vision are formulated as intricate energy minimization problems. In this thesis, motion analysis in image sequences is thoroughly investigated, showing that splitting an original complex problem into simplified sub-problems yields improved accuracy, increased robustness, and a clear and accessible approach to state-of-the-art motion estimation techniques. In Chapter 4, optical flow is considered. Optical flow is commonly estimated by minimizing the combined energy, consisting of a data term and a smoothness term. These two parts are decoupled, yielding a novel and iterative approach to optical flow. The derived Refinement Optical Flow framework is a clear and straight-forward approach to computing the apparent image motion vector field. Furthermore this results currently in the most accurate motion estimation techniques in literature. Much as this is an engineering approach of fine-tuning precision to the last detail, it helps to get a better insight into the problem of motion estimation. This profoundly contributes to state-of-the-art research in motion analysis, in particular facilitating the use of motion estimation in a wide range of applications. In Chapter 5, scene flow is rethought. Scene flow stands for the three-dimensional motion vector field for every image pixel, computed from a stereo image sequence. Again, decoupling of the commonly coupled approach of estimating three-dimensional position and three dimensional motion yields an approach to scene ow estimation with more accurate results and a considerably lower computational load. It results in a dense scene flow field and enables additional applications based on the dense three-dimensional motion vector field, which are to be investigated in the future. One such application is the segmentation of moving objects in an image sequence. Detecting moving objects within the scene is one of the most important features to extract in image sequences from a dynamic environment. This is presented in Chapter 6. Scene flow and the segmentation of independently moving objects are only first steps towards machine visual kinesthesia. Throughout this work, I present possible future work to improve the estimation of optical flow and scene flow. Chapter 7 additionally presents an outlook on future research for driver assistance applications. But there is much more to the full understanding of the three-dimensional dynamic scene. This work is meant to inspire the reader to think outside the box and contribute to the vision of building perceiving machines.