Information processor, cloud platform, information processing method, and computer program product thereof

A cloud platform comprising a plurality of first information processors and a second information processor, wherein the second information processor is configured to divide input image data for the first information processors so as to perform a super resolution process as a distributed processing, the super resolution process generating a high resolution image based on a plurality of low resolution input images, and the second information processor comprises:

• an image analyzer that analyzes a characteristic of input image data of a sequence of image frames;
• a division point determining unit that determines, based on a result of analysis by the image analyzer, a criterion to determine a division point for the input image data among a plurality of criteria, including a scene change point, a group of pictures boundary, corresponding points, or a global motion vector, corresponding to a plurality of characteristics of the input image data, and that determines the division point based on the determined criterion;
• a division processor that divides, for each of the first information processors, the input image data at the division point, and allocates each divided input image to each corresponding one of the first information processors;
• and
• a combination processor that combines results of the super resolution process performed by the each of the first information processors, and outputs a result of the combination.

An information processing method performed by an information processor that divides input image data for a plurality of other information processors so as to perform a super resolution process as a distributed processing, the super resolution process generating a high resolution image based on a plurality of low resolution input images, the information processing method comprising:

• analyzing, by an image analyzer, that analyzes a characteristic of input image data of a sequence of image frames;
• first determining, by a division point determining unit and based on a result of the analyzing, a criterion to determine a division point for the input image data among a plurality of criteria, including a scene change point, a group of pictures boundary, corresponding points, or a global motion vector, corresponding to a plurality of characteristics of the input image data;
• second determining, by the division point determining unit, the division point based on the criterion determined at the first determining;
• dividing, by a division processor and for each of the other information processors, the input image data at the division point, and allocating each divided input image to each corresponding one of the other information processors;
• and combining, by a combination processor, results of the super resolution process performed by the each of the other information processors, and outputting a result of the combination.

A computer program product having a non-transitory computer readable medium including programmed instructions used in an information processor that divides input image data for a plurality of other information processors, the other processors performing a super resolution process on the divided input image creating results, the super resolution process generating a high resolution image based on a plurality of low resolution input images, wherein the instructions, when executed by a computer, cause the computer to perform:

• analyzing, by an image analyzer, that analyzes a characteristic of input image data of a sequence of image frames;
• first determining, by a division point determining unit and based on a result of the analyzing, a criterion to determine a division point for the input image data among a plurality of criteria, including a scene change point, a group of pictures boundary, corresponding points, or a global motion vector, corresponding to a plurality of characteristics of the input image data;
• second determining, by the division point determining unit, the division point based on the criterion determined at the first determining;
• dividing, by a division processor and for each of the other information processors, the input image data at the division point, and allocating each divided input image to each corresponding one of the other information processors;
• and
• combining, by a combination processor, results of the super resolution process performed by the each of the other information processors, and outputting a result of the combination.