Displacement vector estimation is one of the open key problems in Computer Vision and Video Coding. For example in Computer Vision the displacement vector estimation is usually the base of some kind of motion estimation. Unfortunately, the displacement vector estimation using statistical methods is always computationally complex what might be a restriction in real-time processing. In this contribution we show that displacement vector estimation can efficiently be performed by using Cellular Neural Networks (CNN). In order to find CNN-templates therefore, we have used the new optimization method of Iterative Annealing.