Abstract
This document proposes a new algorithm for Motion Estimation using the Kalman Filter. The algorithm predicts the motion vectors for the next frame by using the motion vectors of the current frame. The new algorithm is compared with the already existing search algorithms such as Exhaustive search, Three-Step Search, and Four-Step Search and show that the prediction using the Kalman Filter leads to reduction in the computational complexity.
Additional information
Notes on contributors
Manish Bajaj
Manish Bajaj is Senior Undergraduate Student in Electrical Engineering Department, Indian Institute of Technology Delhi. He will be receiving his B. Tech in August, 2012. His research interests include Image and Video Processing, and Wireless communication. E-mail: [email protected]
Brejesh Lall
Brejesh Lall received B.E. degree in electronics and communication engineering from Delhi College of Engineering, Delhi, India in 1991, M.E. in electronics and communication engineering also from Delhi College of Engineering, Delhi, India in 1992 and PhD degree from Indian Institute of Technology Delhi, India in 1999. He joined Hughes Software Systems, Gurgaon, India in 1997 and worked there in the Signal Processing group till 2005. In 2005 he joined Indian Institute of Technology Delhi as an Assistant Professor in the Electrical Engineering Department and is currently working there as an Associate Professor. His research interests include Multirate Signal Processing, Image Processing and Wireless communication. E-mail: [email protected]