Transforming data to create layouts

A computer storage medium having computer readable instructions stored thereupon that, when executed by a computer, cause the computer to:

• obtain data comprising visual content and a content item;
• analyze the visual content to identify at least one selected from a group consisting of
• a salient region of the visual content, the salient region comprising a subject of the visual content, and
• an invariant region of the visual content, the invariant region comprising a region of the visual content wherein a variation of at least one feature is below a threshold to achieve a contrast between the content item and the visual content;
• automatically generate a design to generate a layout based at least partially on at least one rule, the design comprising a viewport frame and at least one of a salient region placement frame and a content placement frame, the at least one rule specifying at least one selected from a group consisting of an aspect ratio of the viewport frame, a size of the viewport frame, a location of the salient region placement frame, a size of the salient region placement frame, a location of the content placement frame, and a size of the content placement frame;
• generate the layout based on the design;
• prompt a user for feedback on the generated design;
• receive feedback from the user for the generated design;
• and
• generate a new design to generate a new layout based at least partially on the feedback,
• wherein prompting the user for the feedback on the generated design includes prompting the user to provide a ranking for at least one selected from a group consisting of the layout, text placement, image position, element location, and image size of the generated design.

An apparatus comprising:

• a processor;
• a memory communicatively coupled to the processor;
• and
• an application program which executes in the processor from the memory and which, when executed by the processor, causes the processor to:
• obtain data comprising visual content and a content item;
• analyze the visual content to identify at least one selected from a group consisting of
• a salient region of the visual content, the salient region comprising a subject of the visual content, and
• an invariant region of the visual content, the invariant region comprising a region of the visual content wherein a variation of at least one feature is below a threshold to achieve a contrast between the content item and the visual content;
• compare the data to a plurality of designs to provide a comparison of the data;
• and
• in response to failing to select one of the plurality of designs based at least partially on the comparison of the data,
• automatically generate a design to generate a layout based at least partially on at least one rule, the design comprising a viewport frame and at least one of a salient region placement frame and a content placement frame, the at least one rule specifying at least one selected from a group consisting of an aspect ratio of the viewport frame, a size range of the viewport frame, a location of the salient region placement frame, a size of the salient region placement frame, a location of the content placement frame, and a size of the content placement frame,
• generate the layout based on the design;
• prompt a user for feedback on the generated design;
• receive feedback from the user for the generated design;
• and
• generate a new design to generate a new layout based at least partially on the feedback,
• wherein prompting the user for the feedback on the generated design includes displaying a “dislike” prompt for at least one selected from a group consisting of the layout, text placement, image position, element location, and image size of the generated design, and wherein receiving the feedback from the user for the generated design includes receiving a selection of the “dislike” prompt.