Face equalization in video conferencing
View Patent ↗A video conferencing device, which includes a processor, a memory, and a interface configured to receive a video stream having a plurality of objects. The video conferencing device also includes an object selector configured to select a region surrounding a first object in the plurality of objects in the video stream, wherein the first object has abnormal characteristics when compared to a second object, a transformation generator configure to create a transformation, and a transformation module configured to apply the transformation to the video stream to create a transformed video stream, in which the first object appears in a manner more consistent with the second object.
1. A video conferencing device having object equalization comprising:
a processor;
a memory;
an interface configured to receive a video stream having a plurality of objects;
an object selector configured to select a region surrounding a first object in a plurality of objects in the video stream, wherein the first object has abnormal characteristics when compared to a second object in the plurality of objects;
a transformation generator configured to:
create a transformation;
associate the transformation with the first object; and
a transformation module configured to:
determine whether the first object appears in a portion of the video; stream;
when the first object appears in the portion of the video stream, apply the associated transformation to the video stream to create a transformed video stream, in which the first object appears in a manner more consistent with the second object.
2. The video conferencing device of claim 1 , further comprising a microphone.
3. The video conferencing device of claim 1 , further comprising a display.
4. The video conferencing device of claim 1 , further comprising a camera configured to generate the video stream.
5. The video conferencing device of claim 1 , wherein the region is a predetermined number of pixels larger or smaller than the first object.
6. The video conferencing device of claim 1 , wherein the object selector is further configured to:
receive an input, wherein the input defines the region surrounding the first object.
7. The video conferencing device of claim 1 , wherein the object selector is further configured to:
automatically detect the plurality of objects in the video stream;
determine whether the first object has abnormal characteristics when compared to the second object; and
define the region surrounding the first object.
8. The video conferencing device of claim 1 , wherein the transformation is a fish eye transformation.
9. The video conferencing device of claim 1 , wherein the transformation is a luminescence transformation.
10. The video conferencing device of claim 1 , wherein the transformation generator is further configured to:
determine a transformation based upon a set of abnormal characteristics exhibited by an object in the plurality of objects; and
display a dialog confirming whether to apply the transformation to the object.
11. The video conferencing device of claim 4 , wherein the camera is further configured to display the first object when sound is detected from the first object.
12. The device of claim 1 , wherein the video stream is received from a remote device.
13. A method for object equalization in a device comprising:
receiving, at a video conferencing device, a video stream having a plurality of objects;
selecting a region surrounding a first object in the plurality of objects in the video stream, wherein the first object has abnormal characteristics when compared to a second object in the plurality of objects;
creating a transformation;
associate the transformation with the first object;
determine whether the first object appears in a portion of the video stream; and
when the first object appears in the portion of the video stream, applying the associated transformation to the video stream to create a transformed video stream in which the first object appears in a manner more consistent with the second object.
14. The method of claim 13 , wherein selecting the region comprises:
selecting the region in the video stream surrounding the first object, wherein the region is a predetermined number of pixels larger or smaller than the first object.
15. The method of claim 13 , wherein selecting the region comprises:
receiving an input, wherein the input defines the region surrounding the first object.
16. The method of claim 13 , wherein selecting the region comprises:
automatically detecting the plurality of objects in the video stream;
determining whether the first object has abnormal characteristics when compared to the second object; and
defining the region surrounding the first object.
17. The method of claim 13 , wherein the transformation is a fish eye transformation.
18. The method of claim 13 , wherein the transformation is a luminescence transformation.
19. The method of claim 16 , wherein creating a transformation further comprises:
determining the transformation based upon a set of abnormal characteristics exhibited by the first object; and
displaying a dialog confirming whether to apply the transformation to the first object.
20. The method of claim 13 , further comprising:
displaying the first object in the video stream when sound is detected originating from the first object.