Temporal interpolation of video sequences using zonal based algorithms

Temporal interpolation has been recently proposed as a solution for increasing temporal resolution or even for predicting missing or corrupted frames within a video sequence. In this paper new techniques on temporal interpolation are presented, by mainly exploiting properties of the very popular and highly efficient zonal based motion estimation algorithms, and by introducing several other techniques such as multihypothesis motion compensation, motion classification and temporal/spatial filtering. In addition we further give an analysis on when temporal interpolation should be employed, thus possibly avoiding unwanted artifacts created from this process, while at the same time significantly improving the overall performance of the interpolation.

• Type: Conference
• Year: 2001
• Where: ICIP
• Authors: Alexis M. Tourapis, Hye-Yeon Cheong, Ming L. Liou, Oscar C. Au
• Keywords: ICIP 2001, Image Processing, Motion Estimation Algorithms, Temporal Interpolation, Temporal Resolution