Color correction for video communications using display content color information

A method comprising:

• outputting, with a video conference application, display content on a display;
• capturing an image of a user at a location;
• identifying an area of interest (AOI) within the image, wherein the AOI identified includes at least a portion of the face of the user;
• determining whether the user is a known user, wherein the video conference application maintains facial color information for each of the known users;
• determining, based on color information for the display content and on an ambient light measurement made by an ambient light sensor at the location, the extent to which the area of interest was illuminated by light from the display during capture of the image;
• and
• responsive to determining that the area of interest was illuminated by light from the display, performing color correction on the image, wherein performing color correction includes adjusting a color within the area of interest based on the measured ambient light, the extent to which the area of interest was illuminated by light from the display during capture of the image, and, if the user is a known user, the facial color information of the known user.

A non-transitory, computer-readable medium comprising instructions that, when executed, cause one or more processors to:

• output, with a video conference application, display content on a display;
• capture an image of a user at a location;
• identify an area of interest (AOI) within the image, wherein the AOI identified includes at least a portion of the face of the user;
• determine, from the AOI, whether the user is a known user, wherein the video conference application maintains facial color information for each of the known users;
• determine, based on color information for the display content and on an ambient light measurement made by an ambient light sensor at the location, that the area of interest was illuminated by light from the display during capture of the image data;
• and
• responsive to determining that the area of interest was illuminated by light from the display, perform color correction on the image, wherein performing color correction includes adjusting a color within the area of interest based on the measured ambient light, the extent to which the area of interest was illuminated by light from the display during capture of the image, and, if the user is a known user, the facial color information for the known user.

A video presence system, comprising:

• a display;
• a camera;
• an ambient light sensor;
• one or more processors, the one or more processors connected to the display, the camera and the ambient light sensor, the processors configured to output display content on a display;
• memory connected to the one or more processors, wherein the memory includes facial color information for one or more known users;
• and
• a color correction unit connected to the ambient light sensor, wherein the color correction unit receives an image of a user at a location, determines if the user is one of the known users, and corrects color within the image before transmitting the color corrected image, and wherein the color correction includes a first correction based on an ambient light measurement made by the ambient light sensor at the location and a second correction based on display light associated with the display content, wherein the color correction unit is configured to adjust a color of the face of the user within the area of interest based on the ambient light measurement, the extent to which the area of interest was illuminated by light from the display during capture of the image, and, if the user is one of the known users, the facial color information for the known user.

A method comprising:

• detecting one or more known faces in image data of a scene captured by a video camera;
• retrieving facial color information for each detected known face;
• receiving, from an ambient light sensor, a measurement of ambient light falling on the scene;
• determining chromaticity in display content output on a display at the scene during capture of the image data;
• analyzing color of the image data for each detected known face to determine, for each respective detected known face, based on the chromaticity of the received display content and facial skin color characteristics for each respective detected known face, whether the known face detected in the scene is affected by light from a display displaying the display content;
• when any of the known faces detected in the scene are affected by light from a display displaying the display content:
• correcting image data based on the ambient light measurement, the extent to which the area of interest was illuminated by light from the display during capture of the image, and the facial color information retrieved for each respective known user detected in the scene;
• and
• when none of the detected known faces are affected by light from a display displaying the display content, applying ambient light correction across the image.