Method for obtaining image by decode

Abstract

The method comprises: getting the motion vector of relevant block in backward reference frame from bit-stream, which is used for derivation; deciding if the motion vector is over the forward reference frame which is what the B frame maximally appoints to; if not, the calculating the forward and backward motion factor with normal way; if yes, using the motion vector of forward reference frame capable of being obtained by B frame in same direction to replace the motion vector of relevant block of backward reference frame; calculating the forward and backward motion vector of B frame; averaging the pixels corresponding to the image reference block appointed by the forward and backward motion vector to get two-way forecast reference block; adding the two-way forecast reference block to the decoded block residues to get the video image of macro block corresponding to the B frame.

Description

technical field [0001] The present invention relates to a method for decoding and obtaining an image in the field of digital video compression, especially a method for obtaining a reference block of an image in an encoding mode with a fixed number of reference frames, and a method for obtaining the image by decoding and obtaining the reference block of the image, belonging to Image compression technology field. Background technique [0002] Efficient video codec technology is the key to realize high-quality, low-cost multimedia data storage and transmission. The more popular international standards for image coding now use a coding method that combines motion compensation based on block matching, discrete cosine transform and quantization. Typical international standards such as MPEG-1, MPEG-2 and MPEG-4 introduced by the International Organization for Standardization / International Electrotechnical Commission Joint Technical Group 1 (ISO / IEC JTC1), as well as those proposed...

AI Technical Summary

Problems solved by technology

[0010] For P and B frames, when encoding with a fixed number of reference frames, consider that the number of forward reference frames of P frames is more than the number of forward reference frames of B frames (B frames must include a backward reference frame, while P frames do not) , so it will cause the problem of motion vector mismatch, that is: for B frame, when deriving the motion vector, the motion vector of each block in the backward reference frame needs to be used, since the backward reference frame is a P frame, this may It will cause the reference frame pointed by the motion vector to exceed the maximum forward reference frame that the B frame can point to; for example, if figure 1 As shown, when the number of fixed reference frames is 2 frames, the reference frame that the B frame can point to is its two adjacent P frames, and the reference frame of the P frame that is the backward reference frame of the B frame is the previous P frame 2 P frames, that is, P_REF_1 and P_REF_0, when the motion vector of the P frame points to the front P frame P_REF_1, the P frame exceeds the maximum forward reference frame that the B frame can point to, and the calculated motion vector points to The P_REF_1 that the B frame cannot reach, therefore, the B frame cannot obtain the real reference block for encoding, which will make the encoding deviate and cause serious image distortion

Images