Bidirectional motion estimation device for production of interpolated frame based on front reference frame and rear reference frame

Abstract

The invention discloses a bidirectional motion estimation device for production of an interpolated frame based on a front reference frame and a rear reference frame, which comprises a storage unit, a matching error calculation unit, a comparator tree unit and a data memory. The storage unit is used for caching data in the front reference frame and the rear reference frame; the matching error calculation unit is used for dividing the interpolated frame into a plurality of macro blocks of the same size, setting a search range based on each macro block, and calculating matching errors corresponding to all motion vectors in the search range; the comparator tree unit is used for comparing all the matching errors calculated in the search area to obtain the minimum matching error and outputting the minimum matching error and a corresponding motion vector; and the data memory is used for storing the minimum matching error and the corresponding motion vector. The full-search bidirectional motion estimation method can be implemented flexibly, and the efficiency of implementation is high; the driving capacity of signals can be improved, the delay of the signals can be reduced, and the signal synchronization can be controlled; the highest efficiency and data multiplexing can be realized; and the bidirectional motion estimation device has higher flexibility and practicality.

Description

technical field [0001] The invention relates to the technical field of video image processing, in particular to a method for generating an interpolation frame based on two reference frames before and after and a bidirectional motion estimator thereof. [0002] Background technique [0003] Motion estimation algorithm is the core of various existing video image coding technologies, and also the core of video image post-processing technology based on motion estimation and motion compensation (ME / MC). Its performance directly determines the quality of video image processing. It is of great significance to the video processing system. At the same time, the motion estimator is also the module that consumes the most computing resources in a video compression coding chip or a video image post-processing chip. Generally speaking, in a video coding or video image post-processing hardware system, the motion estimator takes up 40% to 50%. hardware overhead. How to improve the operat...

AI Technical Summary

Problems solved by technology

However, for the frame rate increase, the purpose is to obtain a good quality interpolated image, and the problem of unidirectional motion estimation for frame rate increase is that it will produce holes and overlaps in the interpolated frame

However, the application of the above structure is limited due to the design based on unidirectional motion estimation, and the results are not satisfactory for applications such as frame rate improvement

Images