Mixed Reality applications require accurate knowledge of the relative positions of the camera and the scene. When either of them moves, this means keeping track in real-time of all six degrees of freedom that define the camera position and orientation relative to the scene, or, equivalently, the 3D displacement of an object relative to the camera. Many technologies have been tried to achieve this goal. However, Computer Vision is the only one that has the potential to yield non-invasive, accurate and low-cost solutions to this problem, provided that one is willing to invest the effort required to develop sufficiently robust algorithms. In this chapter, we therefore discuss some of the most promising approaches, their strengths, and their weaknesses.