This paper solves the problem to assign optimal sample sizes to Viola's stochastic matching and determines the best stochastic gradient optimizer to this sort of problems. It can be used for applications like X-ray based minimal invasive interventions or the control of patient motion during radiation therapy. The preprocessing for optimally estimating the parameters lies between 0.5-4.5 seconds and is only necessary once for a typical set of images to be matched. Matching itself is performed within 300-1300 milliseconds on an Athlon 800 MHz processor. © Springer-Verlag Berlin Heidelberg 2004.
CITATION STYLE
Müller, U., Hesser, J., & Männer, R. (2004). Fast rigid 2D-2D multimodal registration. In Lecture Notes in Computer Science (Vol. 3216, pp. 887–894). Springer Verlag. https://doi.org/10.1007/978-3-540-30135-6_108
Mendeley helps you to discover research relevant for your work.