Removing reflected information from within a video capture feed during a videoconference
Systems and methods for detecting information displayed by a display device and reflected into a camera feed are disclosed. Described systems and methods remove the reflected information from the camera feed so that unintended disclosure of private information when the camera feed is shared can be prevented.
1. A computer-implementable method for removing reflected information from within a video capture feed, comprising:
displaying, on the display device, a first frame of visual data comprising one or more first images;
capturing, by an image capture device when the display device displays the first frame, a second frame of visual data;
detecting, by the processor in the second frame, a second image corresponding to one or more of the one or more first images;
removing, from the second frame when the image capture device captures the second frame, at least some visual data corresponding to the one or more first images so as to generate a third frame of visual data from the second frame; and
including, for rendering by an electronic device, the third frame within a visual data stream.
2. The computer-implementable method of claim 1 , further comprising:
determining a location of the second image in the second frame;
displaying a fourth frame of visual data using the display device, the fourth frame corresponding to one or more third images;
capturing a fifth frame of visual data using the image capture device; and
searching, by the processor, for one or more fourth images in the fifth frame corresponding to the one or more third images, wherein searching, by the processor, for one or more fourth images in the fifth frame includes initially searching a region of the fifth frame corresponding to the location of the second image in the second frame.
3. The computer-implementable method of claim 1 , further comprising:
transforming the first frame corresponding to one or more first images by applying one or more image transforms to the first frame, wherein searching, by the processor, for one or more second images in the second frame corresponding to the one or more first images comprises searching for one or more second images in the second frame corresponding to one or more transformed first images.
4. The computer-implementable method of claim 3 , wherein transforming the first frame comprises at least one of:
cropping the first frame;
skewing the first frame;
modifying an aspect ratio of the first frame; and
modifying a transparency level of the first frame.
5. The computer-implementable method of claim 3 , wherein transforming the first frame comprises at least one of:
changing a saturation level of the first frame;
changing a blur level of the first frame; and
changing a brightness level of the first frame.
6. The computer-implementable method of claim 1 , wherein including the third frame within a visual data stream for rendering by an electronic device comprises transmitting the audio-visual data stream to a remote endpoint.
7. The computer-implementable method of claim 1 , wherein including the third frame within a visual data stream for rendering by an electronic device comprises displaying at least some of the third image frame using the display device.
8. The computer-implementable method of claim 1 , wherein searching, by the processor, for one or more second images in the second frame corresponding to one or more of the one or more first images comprises searching for image data depicting an eyeglass and wherein detecting, by the processor, a second image in the second frame corresponding to the one or more first images comprises determining that the second image is at least partially bounded by image data depicting the eyeglass.
9. A teleconferencing system, comprising:
a display device;
an image capture device configured to capture a series of image data frames;
a processor coupled to the image capture device and the display device, and configured to process the series of image data frames; and
a non-transitory computer-readable storage medium storing instructions executable by the processor, the instructions when executed by the processor causes the teleconferencing system to:
display a first frame of visual data on the display device, the first frame comprising one or more first images;
capture, by the image capture device when the display device displays the first frame, a second frame of visual data;
detect, in the second frame, a second image corresponding to the one or more first images;
remove, from the second frame when the image capture device captures the second frame, at least some visual data corresponding to the one or more first images so as to generate a third frame of visual data from the second frame; and
include, for rendering by an electronic device, the third frame within a visual data stream.
10. The teleconferencing system of claim 9 , wherein the instructions further comprise instructions to:
determine a location of the second image in the second frame;
display a fourth frame of visual data using the display device, the fourth frame corresponding to one or more third images;
capture a fifth frame of visual data using the image capture device; and
search for one or more fourth images in the fifth frame corresponding to the one or more third images, wherein searching, by the processor, for one or more fourth images in the fifth frame includes initially searching a region of the fifth frame corresponding to the location of the second image in the second frame.
11. The teleconferencing system of claim 10 , wherein the instructions further comprise instructions to:
transform the first frame corresponding to one or more first images by applying one or more image transforms to the first frame, and wherein
the instructions to search for one or more second images in the second frame corresponding to the one or more first images comprises searching for one or more second images in the second frame corresponding to one or more transformed first images.
12. The teleconferencing system of claim 11 , wherein the instructions to transform the first frame comprise instructions to:
crop the first frame;
skew the first frame;
modify an aspect ratio of the first frame; and
modify a transparency level of the first frame.
13. The teleconferencing system of claim 11 , wherein the instructions to transform the first frame comprise instructions to:
change a saturation level of the first frame;
change a blur level of the first frame; and
change a brightness level of the first frame.
14. The teleconferencing system of claim 12 , wherein including the third frame within a visual data stream for rendering by an electronic device comprises transmitting the audio-visual data stream to a remote endpoint.
15. The teleconferencing system of claim 9 , wherein including the third frame within a visual data stream for rendering by an electronic device comprises displaying at least some of the third image frame using the display device.
16. The teleconferencing system of claim 9 , wherein detecting, by the processor, a second image in the second frame corresponding to the one or more first images comprises determining that the second image is at least partially bounded by image data depicting an eyeglass.
17. A non-transitory computer-readable storage medium storing instructions executable by the processor, the instructions when executed by the processor causes the teleconferencing system to:
display, on the display device, a first frame of visual data comprising one or more first images;
capture, by an image capture device when the display device displays the first frame, a second frame of visual data;
detect, in the second frame, a second image corresponding to the one or more first images;
remove, from the second frame when the image capture device captures the second frame, at least some visual data corresponding to the one or more first images so as to generate a third frame of visual data from the second frame; and
include, for rendering by an electronic device, the third frame within a visual data stream.
18. The non-transitory computer readable medium of claim 17 , wherein the instructions further comprise instructions to:
determine a location of the second image in the second frame;
display a fourth frame of visual data using the display device, the fourth frame corresponding to one or more third images;
capture a fifth frame of visual data using the image capture device; and
search, by the processor, for one or more fourth images in the fifth frame corresponding to the one or more third images, wherein searching, by the processor, for one or more fourth images in the fifth frame includes initially searching a region of the fifth frame corresponding to the location of the second image in the second frame.
19. The non-transitory computer readable medium of claim 17 , wherein the instructions further comprise instructions to:
transform the first frame corresponding to one or more first images by applying one or more image transforms to the first frame, and
wherein searching, by the processor, for one or more second images in the second frame corresponding to the one or more first images comprises searching for one or more second images in the second frame corresponding to one or more transformed first images.
20. The non-transitory computer readable medium of claim 19 , wherein the instructions to transform the first frame comprise instructions to:
skew the first frame;
modify an aspect ratio of the first frame; and
modify a transparency level of the first frame.