Semi-background replacement based on rough segmentation
A telecommunication device includes an image capture system that captures an image of a local participant in a telecommunication session, the image comprising foreground and background images defined by plural pixels, each of the pixels having a pixel magnitude related to a sample of the image at a spatial location of the respective pixel and a background modifier that segments plural pixels of the captured image into foreground and background sets of pixels based on spatial coordinates of the pixels, replaces the background set of pixels with a template set of pixels to form a new background set of pixels, each selected pixel in the template set of pixels having a different magnitude than a magnitude of the corresponding pixel in the background set of pixels replaced by the selected pixel, and combines the new background set of pixels with the foreground set of pixels to form modified image information for transmission to a remote endpoint.
1. A telecommunication device, comprising:
an image capture device capturing an image of a local participant in a telecommunication session, the image comprising foreground and background images defined by plural pixels, each of the plural pixels having a pixel magnitude related to a sample of the image at a spatial location of the respective pixel;
a processor coupled with the image capture device; and
a memory coupled with the processor and storing there a set of instructions which, when executed by the processor, causes the processor to:
receive the captured image from the image capture device;
segment the plural pixels of the captured image into foreground and background sets of pixels based on spatial coordinates of the pixels and independent of pixel magnitudes;
replace the background set of pixels with a template set of pixels to form a new background set of pixels, each selected pixel in the template set of pixels having a different magnitude than a magnitude of the corresponding pixel in the background set of pixels replaced by the selected pixel;
combine the new background set of pixels with the foreground set of pixels to form modified image information; and
provide the modified image information to a remote endpoint and/or to a display to display the modified image information to the local participant.
2. The device of claim 1 , wherein the image capture device is a camera, wherein the pixel magnitude is one or more of a pixel value, color plane, and colormap index of the corresponding pixel, wherein the foreground image comprises pixels defining an image of the local participant, wherein the background image comprises pixels defining one or more background objects, wherein the foreground set of pixels comprises pixels defining the image of the local participant and part of the one or more background objects, and wherein the background set of pixels comprises pixels defining the other part of the one or more background objects.
3. The device of claim 2 , wherein the segmentation is based on a selected boundary dividing the background image into first and second subsets of background image information, pixels in the first subset of background image information being in the foreground set of pixels and pixels in the second subset of background image information being in the background set of pixels, and wherein a spatial position of the boundary being related to a dimension of a detected face of the local participant.
4. The device of claim 3 , wherein a spatial position of the boundary spatially moves across multiple frames based on movement of the local participant, wherein movement of the local participant is tracked by tracking movement of a selected facial feature of the local participant, and wherein the boundary spatially moves only when a degree of spatial displacement of the local participant image from a selected position equals or exceeds a selected threshold.
5. The device of claim 3 , wherein input of the local participant is received to alter a spatial position of the boundary from a first position selected automatically to a second position selected manually by the local participant.
6. The device of claim 1 , wherein the captured image is a single 2-dimensional image.
7. The device of claim 6 , wherein the instructions further cause the processor to modify magnitudes of the pixels at a selected boundary to provide a desired visual effect and wherein at least some of the pixels having modified magnitudes are in the foreground set of pixels and/or at least some of the pixels having modified magnitudes are in the new background set of pixels.
8. A telecommunication device, comprising:
an image capture device capturing a 2-dimensional image of a local participant in a telecommunication session, the image comprising foreground and background images defined by plural pixels, each of the plural pixels having a pixel magnitude related to a sample of the image at a spatial location of the respective pixel;
a processor coupled with the image capture device; and
a memory coupled with and readable by the processor and storing therein a set of instructions which, when executed by the processor, causes the processor to:
receive the captured 2-dimensional image from the image capture device;
segment the plural pixels of the captured image into foreground and background sets of pixels based on spatial coordinates of the pixels;
replace the background set of pixels with a template set of pixels to form a new background set of pixels;
combine the new background set of pixels with the foreground set of pixels to form modified image information; and
provide the modified image information to a remote endpoint and/or to a display to display the modified image information to the local participant.
9. The device of claim 8 , wherein the pixel magnitude is one or more of a pixel value, color plane, and colormap index of the corresponding pixel, wherein the foreground image comprises pixels defining an image of the local participant, wherein the background image comprises pixels defining one or more background objects, wherein the foreground set of pixels comprises pixels defining the image of the local participant and part of the one or more background objects, wherein each selected pixel in the template set of pixels having a different magnitude than a magnitude of the corresponding pixel in the background set of pixels replaced by the selected pixel, and wherein the background set of pixels comprises pixels defining the other part of the one or more background objects.
10. The device of claim 9 , wherein the segmentation is based on a selected boundary dividing the background image into first and second subsets of background image information, pixels in the first subset of background image information being in the foreground set of pixels and pixels in the second subset of background image information being in the background set of pixels, and wherein a spatial position of the boundary being related to a dimension of a detected face of the local participant.
11. The device of claim 10 , wherein a spatial position of the boundary spatially moves across multiple frames based on movement of the local participant, wherein movement of the local participant is tracked by tracking movement of a selected facial feature of the local participant, and wherein the boundary spatially moves only when a degree of spatial displacement of the local participant image from a selected position equals or exceeds a selected threshold.
12. The device of claim 10 , wherein input of the local participant is received to alter a spatial position of the boundary from a first position selected automatically to a second position selected manually by the local participant.
13. The device of claim 8 , wherein the segmentation of the plural pixels into foreground and background sets of pixels is independent of pixel magnitudes and wherein the captured image is a single 2-dimensional image.
14. The device of claim 13 , wherein the instructions cause the processor to modify magnitudes of the pixels at a selected boundary to provide a desired visual effect and wherein at least some of the pixels having modified magnitudes are in the foreground set of pixels and/or at least some of the pixels having modified magnitudes are in the new background set of pixels.
15. A method, comprising:
automatically capturing, by an image capture device, an image of a local participant in a telecommunication session, the image comprising foreground and background images defined by plural pixels, each of the plural pixels having a pixel magnitude related to a sample of the image at a spatial location of the respective pixel;
automatically segmenting, by a microprocessor and in substantial real time with image capture, the plural pixels of the captured image into foreground and background sets of pixels based on spatial coordinates of the pixels and independent of pixel magnitudes;
automatically replacing, by the microprocessor, the background set of pixels with a template set of pixels to form a new background set of pixels, each selected pixel in the template set of pixels having a different magnitude than a magnitude of the corresponding pixel in the background set of pixels replaced by the selected pixel;
automatically combining, by the microprocessor, the new background set of pixels with the foreground set of pixels to form modified image information; and
providing, by an output, the modified image information to a remote endpoint and/or to a display to display the modified image information to the local participant.
16. The method of claim 15 , wherein the image capture device is a camera, wherein the pixel magnitude is one or more of a pixel value, color plane, and colormap index of the corresponding pixel, wherein the foreground image comprises pixels defining an image of the local participant, wherein the background image comprises pixels defining one or more background objects, wherein the foreground set of pixels comprises pixels defining the image of the local participant and part of the one or more background objects, and wherein the background set of pixels comprises pixels defining the other part of the one or more background objects.
17. The method of claim 16 , wherein the segmentation is based on a selected boundary dividing the background image into first and second subsets of background image information, pixels in the first subset of background image information being in the foreground set of pixels and pixels in the second subset of background image information being in the background set of pixels, and wherein a spatial position of the boundary being related to a dimension of a detected face of the local participant.
18. The method of claim 17 , wherein a spatial position of the boundary spatially moves across multiple frames based on movement of the local participant, wherein movement of the local participant is tracked by tracking movement of a selected facial feature of the local participant, and wherein the boundary spatially moves only when a degree of spatial displacement of the local participant image from a selected position equals or exceeds a selected threshold.
19. The method of claim 15 , wherein the captured image is a single 2-dimensional image.
20. The method of claim 19 , wherein the microprocessor modifies magnitudes of the pixels at a selected boundary to provide a desired visual effect and wherein at least some of the pixels having modified magnitudes are in the foreground set of pixels and/or at least some of the pixels having modified magnitudes are in the new background set of pixels.