Map converting method, map converting apparatus, and computer program product for map conversion
View Patent ↗According to an embodiment, a map converting method includes calculating existing frequency of depths in a first map, the first map corresponding to at least an image area of an image and each pixel of the first map representing a depth corresponding to a pixel of the image area; and first converting the first map into a second map that represents a depth in a first range for each pixel by using the existing frequency.
1. A map converting method, comprising:
obtaining a first map, the first map corresponding to a particular image and representing depths of pixels in at least part of the particular image;
calculating existing frequencies of depths in the first map and a cumulative histogram that represents cumulative existing frequencies of the depths of the first map;
first converting the first map into a second map by using the cumulative histogram, each depth in the second map being within a first range;
second converting the first map into a third map that represents the depth in the first range for each pixel by a basic conversion with which an entire depth represented by the first map is evenly converted, each depth in the third map being within the first range; and
generating a fourth map by combining the second map multiplied by a blend constant with the third map multiplied by a complement of the blend constant, wherein
the blend constant is a predetermined value.
2. The method according to claim 1 , wherein, in the first converting,
the depth for each pixel represented by the first map is converted into a depth of an allocated range that becomes larger in the second map as the existing frequencies are higher, and is converted into a depth of an allocated range that becomes smaller in the second map as the existing frequencies are lower.
3. The method according to claim 1 , wherein the blend constant is a fixed predetermined value.
4. The method according to claim 1 , wherein the blend constant is determined by a user.
5. The method according to claim 1 , wherein the first map is calculated by using pixel values in at least part of the particular image.
6. The method according to claim 1 , wherein the first range is determined according to a display device to which the fourth map is outputted.
7. The method according to claim 1 , wherein
the calculating includes calculating the existing frequencies for the first map that represent a depth for each pixel of the at least part of the particular image based on a depth value of a second range.
8. The method according to claim 1 , wherein
the calculating includes calculating the existing frequencies for the first map that represent a depth for each pixel of the at least part of the particular image based on a parallax of a second range.
9. A map converting apparatus, comprising:
a map generator configured to obtain a first map, the first map corresponding to a particular image and representing depths of pixels in at least part of the particular image;
a calculator configured to calculate existing frequencies of depths in the first map and a cumulative histogram that represents cumulative existing frequencies of the depths of the first map;
a first converter configured to convert the first map into a second map by using the cumulative histogram, each depth in the second map beginning within a first range;
a second converter configured to convert the first map into a third map by using a linear function with which an entire depth in the first map is evenly converted, each depth in the third map being within the first range; and
a third converter configured to convert the second and third maps to a fourth map by combining the second map multiplied by a blend constant with the third map multiplied by a complement of the blend constant, wherein
the blend constant is a predetermined value.
10. A computer program product comprising a non-transitory computer-readable medium including programmed instructions for map converting, wherein the instructions, when executed by a computer, cause the computer to perform:
obtaining a first map, the first map corresponding to a particular image and representing depths of pixels in at least part of the particular image;
calculating existing frequencies of depths in the first map and a cumulative histogram that represents cumulative existing frequencies of the depths of the first map;
first converting the first map into a second map by using the cumulative histogram, each depth in the second map being within a first range;
second converting the first map into a third map that represents the depth in the first range for each pixel by a basic conversion with which an entire depth represented by the first map is evenly converted, each depth in the third map being within the first range; and
generating a fourth map by combining the second map multiplied by a blend constant with the third map multiplied by a complement of the blend constant, wherein
the blend constant is a predetermined value.