Image generator and method of image generation
Provided is an information-processing device including: a CPU; and a memory storing instructions for causing the information-processing device, when executed by the CPU, to: output an intermediate heatmap for input of an input image by using at least one of a plurality of machine learning models; and generate a heatmap based on an attribute of the input image, which is provided independently of the input image, and the intermediate heatmaps.
1 . An information-processing device, comprising:
a CPU; and
a memory storing instructions for causing the information-processing device, when executed by the CPU, to:
output an intermediate heatmap for input of an input image by using at least one of a plurality of machine learning models;
generate a heatmap based on an attribute of the input image and the intermediate heatmaps;
wherein the attribute of the input image is provided independently of the input image; and
wherein each of the plurality of machine learning models is configured to output the intermediate heatmap irrespective of any other machine learning models among the plurality of machine learning models.
2 . The information-processing device according to claim 1 , wherein the instructions further cause the information-processing device to:
select, as a machine learning model to which the input image is to be input, at least one machine learning model from the plurality of machine learning models based on the attribute.
3 . The information-processing device according to claim 1 , wherein the instructions further cause the information-processing device to:
select at least one intermediate heatmap from a plurality of the intermediate heatmaps output from the plurality of machine learning models based on the attribute.
4 . The information-processing device according to claim 1 , wherein the instructions further cause the information-processing device to:
generate the heatmap by giving a weight to each of a plurality of the intermediate heatmaps and combining the plurality of the weighted intermediate heatmaps.
5 . The information-processing device according to claim 4 , wherein the instructions further cause the information-processing device to:
determine at least a part of the weights based on the attribute.
6 . The information-processing device according to claim 1 , wherein the instructions further cause the information-processing device to:
cut out a principal portion being a part of the input image based on the heatmap.
7 . The information-processing device according to claim 1 , wherein the plurality of machine learning models includes at least one of a first machine learning model that outputs a click through rate prediction or a second machine learning model that outputs aesthetic values.
8 . The information-processing device according to claim 1 , wherein each of the plurality of machine learning models is configured to output a heatmap independent of the attribute.
9 . The information-processing device according to claim 2 , wherein at least one of the unselected machine learning models from the plurality of machine learning models does not generate a heat map.
10 . The information-processing device according to claim 3 , wherein only a subset of the plurality of generated intermediate heat maps are selected and used to generate the heatmap.
11 . The information-processing device according to claim 6 , wherein the instructions cause the information-processing device to:
cut out the principal portion using a sliding window by setting a plurality of windows; extracting a plurality of candidate windows among the plurality of windows; and selecting a candidate window among the plurality of candidate windows as the principal portion.
12 . The information-processing device according to claim 6 , wherein the instructions cause the information-processing device to:
cut out the principal portion using a machine learning model to directly output a size, a shape, and a position of the principal portion.
13 . The information-processing device according to claim 1 , wherein the heatmap is an evaluation image.
14 . The information-processing device according to claim 1 , wherein a resolution of the heatmap does not match a resolution of the input image.
15 . The information-processing device according to claim 1 , wherein the heatmap is a combination of a plurality of the intermediate heatmaps.
16 . An information-processing method of causing a computer to execute:
outputting, through use of one or a plurality of machine learning models, one or a plurality of intermediate heatmaps for input of an input image;
generating a heatmap based on an attribute of the input image and the one or the plurality of intermediate heatmaps;
wherein the attribute of the input image is provided independently of the input image; and
wherein each of the plurality of machine learning models is configured to output the intermediate heatmap irrespective of any other machine learning models among the plurality of machine learning models.
17 . A non-transitory computer-readable information recording medium storing an information-processing program for causing a computer to:
output, through use of one or a plurality of machine learning models, one or a plurality of intermediate heatmaps for input of an input image;
generate a heatmap based on an attribute of the input image and the one or the plurality of intermediate heatmaps;
wherein the attribute of the input image is provided independently of the input image; and
wherein each of the plurality of machine learning models is configured to output the intermediate heatmap irrespective of any other machine learning models among the plurality of machine learning models.