Method for encoding and decoding video information, a motion compensated video encoder and a coresponding decoder

Abstract

A method for encoding video information is presented, where a piece of current video information is segmented into macroblocks and a certain number of available macroblock segmentations for segmenting a macroblock into blocks is defined. Furthermore, for each available macroblock segmentation at least one available prediction method is defined, each of which prediction methods produces prediction motion coefficients for blocks within said macroblock resulting in a certain finite number of available macroblock-segmentation—prediction-method pairs. For a macroblock, one of the available macroblock-segmentation—prediction-method pairs is selected, and thereafter the macroblock is segmented into blocks and prediction motion coefficients for the blocks within said macroblock are produced using the selected macroblock-segmentation—prediction-method pair. A corresponding decoding method, an encoder and a decoder are also presented.

Description

[0001] The present invention relates to video coding. In particular, it relates to compression of video information using motion compensated prediction. BACKGROUND OF THE INVENTION [0002] A video sequence typically consists of a large number video frames, which are formed of a large number of pixels each of which is represented by a set of digital bits. Because of the large number of pixels in a video frame and the large number of video frames even in a typical video sequence, the amount of data required to represent the video sequence quickly becomes large. For instance, a video frame may include an array of 640 by 480 pixels, each pixel having an RGB (red, green, blue) color representation of eight bits per color component, totaling 7,372,800 bits per frame. Another example is a QCIF (quarter common intermediate format) video frame including 176×144 pixels. QCIF provides an acceptably sharp image on small (a few square centimeters) LCD displays, which are typically available in mo...

AI Technical Summary

Benefits of technology

[0027] An object of the present invention is to provide a method that provides a flexible and versatile motion coefficient prediction for encoding / decoding video information using motion compensation. A further object of the invention is to provide a motion compensated method for coding / decoding video information that provides good performance in terms of transmission bandwidth and image quality while being computationally fairly simple. A further object is to present a method for encoding / decoding video information that provides satisfactory results when a comparatively simple motion model, such as the translational motion model, is used.

[0038] By restricting the number of possible prediction methods per macroblock segmentation, the complexity of the encoding process is reduced compared, for example, to an encoding process where the best prediction motion coefficient candidate is determined freely using any neighboring blocks or combinations thereof. In such a case, there is a large number of prediction motion coefficient candidates. When the prediction blocks are defined beforehand for each prediction method and there is a limited number of prediction methods per macroblock segmentation, it is possible to estimate the cost of each macroblock-segmentation—prediction-method pair. The pair minimizing the cost can then be selected.

[0039] Advantageously, there is only one available prediction method per macroblock segmentation. This reduces the complexity of the encoding method even further. Furthermore, in this situation it is possible to conclude the prediction method of a block directly from the selected macroblock segmentation. There is thus necessarily no need to transmit information about the prediction method to the decoding entity. Thus, in this case the amount of transmitted information is not increased by adding adaptive features, i.e. various prediction methods used within a frame, to the encoded information.

[0040] By selecting the available prediction blocks and defining the macroblock-segmentation-specific prediction methods suitably, it is possible to implement a high performance video encoding method using at most three predetermined prediction blocks to produce prediction motion coefficients and allowing only one prediction method per macroblock segmentation. For each macroblock, the macroblock-segmentation—prediction-method pair minimizing a cost function is selected. The simple adaptive encoding of motion information provided by the invention is efficient in terms of computation and in terms of the amount of transmitted information and further more yields good image quality.

Problems solved by technology

Because of the large number of pixels in a video frame and the large number of video frames even in a typical video sequence, the amount of data required to represent the video sequence quickly becomes large.

Due to the very large number of pixels in a frame it is not efficient to transmit a separate motion vector for each pixel to the decoder.

Because of the error introduced by quantization, this operation usually produces some degradation in the prediction error frame En(x,y).

The quadratic motion model provides good prediction performance, but it is less popular in coding than the affine model, since it uses more motion coefficients, while the prediction performance is not substantially better than, for example, that of the affine motion model.

Furthermore, it is computationally more costly to estimate the quadratic motion than to estimate the affine motion.

The segmentation of the current frame into segments Sk can, for example, be carried out in such a way that each segment corresponds to a certain object moving in the video sequence, but this kind of segmentation is a very complex procedure.

The main drawback of this method is that finding the best prediction candidate among the already transmitted image segments is a complex task: the encoder has to perform exhaustive calculations to evaluate all the possible prediction candidates and then select the best prediction block.

For example, wireless mobile terminals have limited space for additional components and as they operate by battery, they typically cannot provide computing capacity comparable to that of desktop computers.

Images