Hybrid-resolution encoding and decoding method and a video apparatus using the same
地域：
Shenzhen
摘要
The present invention provides a hybrid-resolution encoding and decoding method. The method includes performing full-resolution standard coding on an I frame that adopts only intra-frame coding in a video frame sequence, and performing frame reconstruction to obtain a reconstructed frame of the I frame; performing down-sampling on the reconstructed frame of the I frame to obtain a first sampling image of the reconstructed frame of the I frame; performing down-sampling on a non-I frame in the video frame sequence to obtain a second sampling image of the non-I frame; performing standard coding on the second sampling image of the non-I frame by using the first sampling image of the reconstructed frame of the I frame that corresponds to the non-I frame as a reference frame; and sending a video code stream that includes the full-resolution standard coded I frame and the standard coded non-I frame.
說明書
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
This application is a continuation application of PCT Patent Application No. PCT/CN2013/075965, entitled “A HYBRID-RESOLUTION ENCODING AND DECODING METHOD AND A VIDEO APPARATUS USING THE SAME” filed on May 21, 2013, which claims priority to Chinese Patent Application No. 201210450379.7, entitled “A HYBRID-RESOLUTION ENCODING AND DECODING METHOD AND A VIDEO APPARATUS USING THE SAME,” filed on Nov. 12, 2012, both of which are incorporated by reference in their entirety.
The disclosed implementations relate generally to the field of video encoding and decoding, and in particular, to a hybrid-resolution coding and decoding method and a video apparatus using the same.
權(quán)利要求
What is claimed is:1. A hybrid-resolution decoding method, comprising:at a video decoding apparatus having memory and one or more processors:receiving a video code stream; performing full-resolution standard decoding on an I frame that adopts only intra-frame coding in the video code stream; performing down-sampling on the I frame obtained from the full-resolution standard decoding to obtain a sampling image of the I frame; performing the standard decoding on a non-I frame in the video code stream by using the sampling image of the I frame that corresponds to the non-I frame as a reference frame, wherein the standard decoding performed on the non-I frame is different from the full-resolution standard decoding performed on the I frame; and performing super-resolution image recovery on the standard decoded non-I frame to obtain an original-resolution non-I frame image, wherein performing super-resolution image recovery on the standard decoded non-I frame further comprises:obtaining a motion vector and residual data of an inter-frame coded macroblock in the standard decoded non-I frame; and, in accordance with a determination, based on the motion vector and the residual data, that an accuracy of motion estimation of the macroblock does not satisfy first accuracy criteria, performing the super-resolution image recovery on the inter-frame coded macroblock using spatial interpolation filtering; and, in accordance with a determination, based on the motion vector and the residual data, that the accuracy of motion estimation of the macroblock satisfies the first accuracy criteria:in accordance with a determination, based on the motion vector and the residual data, that accuracy of motion estimation satisfies second accuracy criteria associated with a movement level of the macroblock, performing the super-resolution image recovery on the inter-frame coded macroblock using temporal interpolation filtering; and, in accordance with a determination, based on the motion vector and the residual data, that the accuracy of motion estimation does not satisfy the second accuracy criteria associated with the movement level of the macroblock: ?in accordance with a determination, based on the motion vector and the residual data, that the accuracy of motion estimation satisfies third accuracy criteria associated with the movement level of the macroblock, performing the super-resolution image recovery on the inter-frame coded macroblock using combined temporal and spatial interpolation filtering; and, ?in accordance with a determination, based on the motion vector and the residual data, that the accuracy of motion estimation does not satisfy the third accuracy criteria associated with the movement level of the macroblock, performing the super-resolution image recovery on the inter-frame coded macroblock using spatial interpolation filtering. 2. The hybrid-resolution decoding method according to claim 1 , wherein performing the super-resolution image recovery on the inter-frame coded macroblock comprises:performing the super-resolution image recovery on the inter-frame coded macroblock by using the full-resolution standard decoded I-frame that corresponds to the non-I frame. 3. The hybrid-resolution decoding method according to claim 1 , wherein performing super-resolution image recovery on the standard decoded non-I frame further comprises:performing the spatial interpolation filtering on an intra-frame coded macroblock in the standard decoded non-I frame. 4. The hybrid-resolution decoding method according to claim 1 , wherein the down-sampling comprises:performing the down-sampling on the full-resolution standard decoded I frame to obtain the sampling image, wherein the width of the sampling image is 1/a of the width of a frame picture, the height of the sampling image is 1/b of the height of the frame picture, and a and b are greater than 1; and if the width and the height of the sampling image are not integral multiples of the width and the height of a coded macroblock of the video code stream, respectively, expanding the sampling image to obtain a second sampling image with the width and the height respectively being an integral multiple of the width and the height of the coded macroblock. 5. The hybrid-resolution decoding method according to claim 1 , wherein:if the non-I frame is a P frame, the corresponding I frame is an I frame prior to the P frame in the video code stream in accordance with a play order; and if the non-I frame is a B frame, the corresponding I frame is an I frame prior to and/or after the B frame in the video code stream in accordance with the play order. 6. A hybrid-resolution decoding apparatus comprising:one or more processors; memory; an a plurality of program modules stored in the memory and to be executed by the one or more processors, the plurality of program modules further including:a code stream receiving module, configured to receive a video code stream; an I frame decoding module, configured to perform full-resolution standard decoding on an I frame that adopts only intra-frame coding in the video code stream; a down-sampling module, configured to perform down-sampling on the full-resolution standard decoded I frame to obtain a sampling image of the I frame; a non-I frame decoding module, configured to perform standard decoding on a non-I frame in the video code stream by using the sampling image of the I frame that corresponds to the non-I frame as a reference frame, wherein the standard decoding performed on the non-I frame is different from the full-resolution standard decoding performed on the I frame; and a super-resolution recovery module, configured to perform super-resolution image recovery on the standard decoded non-I frame to obtain an original-resolution non-I frame image, wherein the super-resolution recovery module comprises:a data obtaining unit, configured to obtain a motion vector and residual data of an inter-frame coded macroblock in the standard decoded non-I frame; and an inter-frame coding recovery unit, configured to:in accordance with a determination, based on the motion vector and the residual data, that an accuracy of motion estimation of the macroblock does not satisfy first accuracy criteria, perform the super-resolution image recovery on the inter-frame coded macroblock using spatial interpolation filtering; and, in accordance with a determination, based on the motion vector and the residual data, that the accuracy of motion estimation of the macroblock satisfies the first accuracy criteria: ?in accordance with a determination, based on the motion vector and the residual data, that accuracy of motion estimation satisfies second accuracy criteria associated with a movement level of the macroblock, perform the super-resolution image recovery on the inter-frame coded macroblock using temporal interpolation filtering; and, ?in accordance with a determination, based on the motion vector and the residual data, that the accuracy of motion estimation does not satisfy the second accuracy criteria associated with the movement level of the macroblock: ?in accordance with a determination, based on the motion vector and the residual data, that the accuracy of motion estimation satisfies third accuracy criteria associated with the movement level of the macroblock, perform the super-resolution image recovery on the inter-frame coded macroblock using combined temporal and spatial interpolation filtering; and ?in accordance with a determination, based on the motion vector and the residual data, that the accuracy of motion estimation does not satisfy the third accuracy criteria associated with the movement level of the macroblock, perform the super-resolution image recovery on the inter-frame coded macroblock using spatial interpolation filtering. 7. The hybrid-resolution decoding apparatus according to claim 6 , wherein performing the super-resolution image recovery on the inter-frame coded macroblock comprises:performing the super-resolution image recovery on the inter-coded macroblock by using the full-resolution standard decoded I frame that corresponds to the non-I frame. 8. The hybrid-resolution decoding apparatus according to claim 6 , wherein the super-resolution recovery module further comprises:an intra-frame coding recovery unit, configured to perform the spatial interpolation filtering on an intra-frame coded macroblock in the standard decoded non-I frame. 9. The hybrid-resolution decoding apparatus according to claim 6 , wherein the down-sampling module comprises:a down-sampling unit, configured to perform the down-sampling on a frame picture of the full-resolution standard decoded I frame to obtain the sampling picture, wherein the width of the sampling picture is 1/a of the width of the frame picture, the height of the sampling picture is 1/b of the height of the frame picture, and a and b are greater than 1; and an expanding unit, configured to expand the sampling image to obtain a second sampling image of the width and the height respectively being an integral multiple of the width and the height of a coded macroblock when the width and the height of the sampling image are not integral multiples of the width and the height of the coded macroblock, respectively. 10. The hybrid-resolution decoding apparatus according to claim 6 , wherein:if the non-I frame is a P frame, the corresponding I frame is an I frame prior to the P frame in the video code stream in accordance with a play order; and if the non-I frame is a B frame, the corresponding I frame is an I frame prior to and/or after the B frame in the video code stream in accordance with the play order.
意見反饋
使用該功能遇到問題