Image segmentation device, image segmentation method, and depth map generating method
An image segmentation method includes displaying, through a display component, an image; receiving, through a user input component, a user command relative to an original designation region of the image; shrinking the original designation region according to first depth information corresponding to the image to generate a shrunk designation region of the image; generating a plurality of seeds according to the shrunk designation region and the user command; and segmenting the image according to the seeds to obtain at least one segmented region.
1. An image segmentation method comprising:
displaying, through a display component, an image;
receiving, through a user input component, a user command relative to an original designation region of the image;
shrinking the original designation region according to first depth information corresponding to the image to generate a shrunk designation region of the image;
generating a plurality of seeds according to the shrunk designation region and the user command; and
segmenting the image according to the seeds to obtain at least one segmented region.
2. The image segmentation method as claimed in claim 1 , wherein the step of shrinking the original designation region according to the first depth information corresponding to the image comprises:
clustering a plurality of image units within the original designation region of the image into two groups according to the first depth information; and
determining the shrunk designation region of the image according to the two groups.
3. The image segmentation method as claimed in claim 2 , wherein the step of clustering the image units within the original designation region of the image into two groups comprises:
assigning each of the image units within the original designation region a weight corresponding to the distances between the image units within the original designation region and a center of the original designation region;
calculating weight-summations by summing the weights corresponding to the image units determined to have an identical depth; and
clustering the image units within the original designation region into two groups according to the weight-summations.
4. The image segmentation method as claimed in claim 2 , wherein the step of determining the shrunk designation region of the image according to the two groups comprises:
selecting a region corresponding to one of the two groups; and
eroding the selected region corresponding to the selected group from an edge of the selected region to generate the shrunk designation region of the image.
5. The image segmentation method as claimed in claim 2 further comprising:
dividing a plurality of pixels of the image into a plurality of superpixels according to colors of the pixels; and
averaging a plurality of depth values corresponding to the pixels within each of the superpixels to generate a plurality of average depth values corresponding to the superpixels;
wherein the superpixels serve as the image units, and the first depth information comprises the average depth values corresponding to the superpixels.
6. The image segmentation method as claimed in claim 5 , wherein the step of segmenting the image according to the seeds to obtain at least one segmented region comprises:
performing a segmentation algorithm according to the image, the seeds, and second depth information, wherein the second depth information comprises the average depth values corresponding to the superpixels.
7. The image segmentation method as claimed in claim 2 , wherein a plurality of pixels of the image serve as the image units, and the depth information comprises a plurality of depth values respectively corresponding to the pixels.
8. The image segmentation method as claimed in claim 1 , wherein the step of generating the seeds according to the shrunk designation region and the user command comprises:
taking a first portion of the pixels of the image located at the edge of the original designation region as background seeds; and
taking a second portion of the pixels of the image located within the shrunk designation region as foreground seeds.
9. An image segmentation device comprises:
a display component configured for displaying an image;
a user input component configured for receiving a user command relative to an original designation region of the image; and
a processing component configured for:
shrinking the original designation region according to depth information corresponding to the image to generate a shrunk designation region of the image;
generating a plurality of seeds according to the shrunk designation region and the user command; and
segmenting the image according to the seeds to obtain at least one segmented region.
10. The image segmentation device as claimed in claim 9 , wherein the processing component is further configured for:
clustering a plurality of image units within the original designation region of the image into two groups according to the depth information; and
determining the shrunk designation region of the image according to the two groups.
11. The image segmentation device as claimed in claim 10 , wherein the processing component is further configured for:
assigning each of the image units within the original designation region a weight corresponding to the distances between the image units within the original designation region and a center of the original designation region;
calculating weight-summations by summing the weights corresponding to the image units determined to have an identical depth; and
clustering the image units within the original designation region into two groups according to the weight-summations.
12. The image segmentation device as claimed in claim 10 , wherein the processing component is further configured for:
selecting a region corresponding to one of the two groups; and
eroding the selected region corresponding to the selected group from an edge of the selected region to generate the shrunk designation region of the image.
13. The image segmentation device as claimed in claim 10 , wherein the processing component is further configured for:
dividing a plurality of pixels of the image into a plurality of superpixels according to colors of the pixels; and
averaging a plurality of depth values corresponding to the pixels within each of the superpixels to generate a plurality of average depth values corresponding to the superpixels;
wherein the superpixels serve as the image units, and the depth information comprises the average depth values corresponding to the superpixels.
14. The image segmentation device as claimed in claim 13 , wherein the processing component is further configured for:
performing a segmentation algorithm according to the image, the seeds, and second depth information, wherein the second depth information comprises the average depth values corresponding to the superpixels.
15. The image segmentation device as claimed in claim 10 , wherein a plurality of pixels of the image serve as the image units, and the depth information comprises a plurality of depth values respectively corresponding to the pixels.
16. The image segmentation device as claimed in claim 9 , wherein the processing component is further configured for:
taking a first portion of the pixels of the image located at the edge of the original designation region as background seeds; and
taking a second portion of the pixels of the image located within the shrunk designation region as foreground seeds.