Video image decoding apparatus and video image encoding system

1. A video image decoding apparatus, comprising:

a processor; and

a non-transitory storage medium containing program instructions, execution of which by the processor causes the video image decoding apparatus to provide functions of

a plurality of predicted image generating units each configured to

generate a predicted image according to one of a plurality of different methods,

estimate an image quality of the generated predicted image, and

output the estimated image quality as image quality information, which includes a plurality of image quality values each corresponding to a pixel of the generated predicted image, together with the generated predicted image;

a predicted image combining unit configured to generate a combined predicted image that is a weighted average of the generated predicted images, by, for each pixel of the combined predicted image,

determining a plurality of pixels, each in one of the generated predicted images at a same pixel location as that of said each pixel in the combined predicted image,

weighting a pixel value of each determined pixel by the image quality value corresponding to said each determined pixel, and

summing the weighted pixel values; and

a decoding unit configured to decode an encoded image by using, as side information, the combined predicted image obtained by the predicted image combining unit.

2. The video image decoding apparatus according to claim 1 , wherein the execution of the program instructions by the processor causes the video image decoding apparatus to further provide functions of a key frame decoding unit configured to decode a key frame encoded image produced by encoding a key frame,

wherein the decoding unit decodes a non-key frame encoded image produced by encoding a non-key frame, and

the predicted image generating units generate the predicted images using at least the decoded key frame.

3. The video image decoding apparatus according to claim 2 , wherein

at least one of the predicted image generating units generates a predicted image using a hash provided from a corresponding video image encoding apparatus.

4. The video image decoding apparatus according to claim 3 , wherein

at least one of the predicted image generating units generates a predicted image from decoded key frames at previous and following times, and

at least another of the predicted image generating units generates a predicted image from the hash and one of the decoded key frame at the previous time, the decoded key frame at the following time, and the decoded key frames at the previous and following times.

5. The video image decoding apparatus according to claim 1 , wherein

each of the predicted image generating units estimates the image quality based on each pixel in the predicted image, and outputs the estimated image quality as the image quality information.

6. The video image decoding apparatus according to claim 1 , wherein

the video image decoding apparatus is provided with a hash from a corresponding video image encoding apparatus, and

each of the predicted image generating units generates a hash from the generated predicted image, and generates the image quality information by using a difference between the generated hash and the provided hash as an estimated value of the image quality.

7. The video image decoding apparatus according to claim 1 , wherein

the video image decoding apparatus is provided with a hash from a corresponding video image encoding apparatus, and

each of the predicted image generating units generates an image from a hash from a time used to estimate a motion vector, obtains an evaluation value of the motion vector used to generate the predicted image between the image generated from the hash and the predicted image, and generates the image quality information using the evaluation value as an estimated value of the image quality.

8. The video image decoding apparatus according to claim 7 ,

wherein each of the predicted image generating units calculates, as the evaluation value of the motion vector, one of a sum of absolute differences between pixel values of each pixel and a mean of absolute differences between pixel values of each pixel between the image generated from the hash and the predicted image.

9. A video image decoding apparatus, comprising:

a processor; and

a non-transitory storage medium containing program instructions, execution of which by the processor causes the video image decoding apparatus to provide functions of

a plurality of predicted image generating units each configured to

generate a predicted image according to one of a plurality of different methods,

estimate an image quality of the generated predicted image, and

output the estimated image quality as image quality information, which includes a plurality of image quality values each corresponding to a pixel of the generated predicted image, together with the generated predicted image;

a predicted image combining unit configured to generate a combined predicted image by, for each pixel of the combined predicted image,

determining a plurality of pixels, each in one of the generated predicted images at a same pixel location as that of said each pixel in the combined predicted image,

obtaining on pixel, among the plurality of determined pixels, that has a highest image quality value corresponding thereto based on the image quality information, and

setting a pixel value of the obtained pixel as a pixel value of said each pixel of the combined predicted image, and

a decoding unit configured to decode an encoded image by using, as side information, the combined predicted image obtained by the predicted image combining unit.

10. The video image decoding apparatus according to claim 9 ,

wherein each of the predicted image generating units estimates the image quality based on each pixel in the predicted image, and outputs the estimated image quality as the image quality information.

11. The video image decoding apparatus according to claim 9 ,

wherein the video image decoding apparatus is provided with a hash from a corresponding video image encoding apparatus, and

each of the predicted image generating units generates a hash from the generated predicted image, and generates the image quality information by using a difference between the generated hash and the provided hash as an estimated value of image quality.

12. The video image decoding apparatus according to claim 9 ,

wherein the video image decoding apparatus is provided with a hash from a corresponding video image encoding apparatus, and

each of the predicted image generating units generates an image from a hash from a time used to estimate a motion vector, obtains an evaluation value of the motion vector used to generate the predicted image between the image generated from the hash and the predicted image, and generates the image quality information using the evaluation value as an estimated value of image quality.

13. The video image decoding apparatus according to claim 12 ,

wherein each of the predicted image generating units calculates, as the evaluation value of the motion vector, one of a sum of absolute differences between pixel values of each pixel and a mean of absolute differences between pixel values of each pixel between the image generated from the hash and the predicted image.

14. A video image encoding system, comprising:

the video image decoding apparatus according to claim 1 ; and

a corresponding video image encoding apparatus.