Systems and method for virtual reality video conversion and streaming

A method comprising:

• accessing, by one or more processors, an image in which a first region is adjacent to second and third regions that are non-adjacent to each other within the image;
• generating, by the one or more processors, a complexity map based on the image that includes the first, second, and third regions;
• based on the generated complexity map and by the one or more processors, determining corresponding dimensions to which the non-adjacent second and third regions of the image are to be scaled;
• scaling, by the one or more processors, the non-adjacent second and third regions of the image to the corresponding dimensions determined based on the complexity map generated based on the image;
• and
• generating, by the one or more processors, a modified version of the image, the modified version including the first region and including the scaled second region positioned adjacent to the scaled third region, the adjacent second and third regions in the modified version of the image being scaled to the corresponding dimensions determined based on the complexity map generated based on the image.

A non-transitory machine-readable storage medium comprising instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:

• accessing an image in which a first region is adjacent to second and third regions that are non-adjacent to each other within the image;
• generating a complexity map based on the image that includes the first, second, and third regions;
• based on the generated complexity map, determining corresponding dimensions to which the non-adjacent second and third regions of the image are to be scaled;
• scaling the non-adjacent second and third regions of the image to the corresponding dimensions determined based on the complexity map generated based on the image;
• and
• generating a modified version of the image, the modified version including the first region and including the scaled second region positioned adjacent to the scaled third region, the adjacent second and third regions in the modified version of the image being scaled to the corresponding dimensions determined based on the complexity map generated based on the image.