Title :
Motion Adaptive Deinterlacing With Modular Neural Networks
Author :
Choi, Hyunsoo ; Lee, Chulhee
Author_Institution :
Dept. of Electr. & Electron. Eng., Yonsei Univ., Seoul, South Korea
fDate :
6/1/2011 12:00:00 AM
Abstract :
In this letter, a motion adaptive deinterlacing algorithm based on modular neural networks is proposed. The proposed method uses different neural networks based on the amount of motion. Modular neural networks were selectively used depending on the differences between the adjacent fields. We also used motion vectors to select optimal input pixels from the adjacent fields. Motion estimation was used to find input blocks for the neural networks with minimum errors. Intra/inter-mode switching was employed to address inaccurate motion estimation problems.
Keywords :
motion estimation; neural nets; inter-mode switching; intra-mode switching; modular neural networks; motion adaptive deinterlacing; motion estimation; motion vectors; Artificial neural networks; Interpolation; Materials; Motion estimation; PSNR; Pixel; Video sequences; Modular neural networks; motion adaptive deinterlacing; motion estimation; neural networks;
Journal_Title :
Circuits and Systems for Video Technology, IEEE Transactions on
DOI :
10.1109/TCSVT.2011.2129190