Scene background blurring including determining a depth map
AssigneeDigitalOptics Corporation Europe Limited
InventorsGuy Michrowski, Gal Shabtay, Noy Cohen, Eyal Ben-Eliezer, Ephraim Goldenberg
Life of this patent
Estimating the term…
Abstract
Different distances of two or more objects in a scene being captured in a video conference are determined using an auto-focus sweep of the scene. A depth map of the scene is generated based on the auto-focus sweep. At least one of the objects is identified as a foreground object or a background object, or one or more of each, based on the determining of the different distances. The technique involves blurring or otherwise rendering unclear at least one background object or one or more portions of the scene other than the at least one foreground object, or combinations thereof, also based on the determining of distances.
Claims48
A method of displaying a participant during a video conference against a blurred or otherwise unclear background, comprising:
- using an imaging device including an optic, an image sensor and a processor;
- determining different distances of two or more objects in a scene being captured in video, including:
- performing an auto-focus sweep of the scene;
- determining a sharpest of two or more color channels, including calculating the sharpest of at least two of three colors:
- sharpest
- =
- {
- j
- |
- σ
- j
- AV
- j
- =
- max
-
- {
- σ
- r
- AV
- r
- ;
- σ
- g
- AV
- g
- ;
- σ
- b
- AV
- b
- }
- }
- where AVi comprise averages of pixels for at least two of three specific color channels {j|r, g, b};
- and
- determining at least two relative distances based on the determining of the sharpest of the two or more color channels;
- generating a depth map of the scene based on the auto-focus sweep;
- and
- identifying at least one of the objects as a foreground object or a background object, or one or more of each, based on the determining of the different distances.
One or more non-transitory computer-readable storage media having code embedded therein for programming a processor to perform a method of displaying a participant during a video conference against a blurred or otherwise unclear background, wherein the method comprises:
- determining different distances of two or more objects in a scene being captured in video, including:
- performing an auto-focus sweep of the scene;
- determining a sharpest of two or more color channels, including calculating the following for at least two of three colors:
- sharpest
- =
- {
- j
- |
- σ
- j
- AV
- j
- =
- max
-
- {
- σ
- r
- AV
- r
- ;
- σ
- g
- AV
- g
- ;
- σ
- b
- AV
- b
- }
- }
- where AVi comprise averages of pixels for at least two of three specific color channels {j|r, g, b};
- and
- determining at least two relative distances based on the determining of the sharpest of the two or more color channels;
- generating a depth map of the scene based on the auto-focus sweep;
- and
- identifying at least one of the objects as a foreground object or a background object, or one or more of each, based on the determining of the different distances.
A video conferencing apparatus, comprising:
- a video camera including a lens, and an image sensor;
- a microphone;
- a display;
- a processor;
- one or more networking connectors;
- and
- one or more computer-readable storage media having code embedded therein for programming a processor to perform a method of displaying a participant during a video conference against a blurred or otherwise unclear background, wherein the method comprises:
- determining different distances of two or more objects in a scene being captured in video, including:
- performing an auto-focus sweep of the scene;
- determining a sharpest of two or more color channels, including calculating the following for at least two of three colors:
- sharpest
- =
- {
- j
- |
- σ
- j
- AV
- j
- =
- max
-
- {
- σ
- r
- AV
- r
- ;
- σ
- g
- AV
- g
- ;
- σ
- b
- AV
- b
- }
- }
- where AVi comprise averages of pixels for at least two of three specific color channels {j|r, g, b};
- and
- determining at least two relative distances based on the determining of the sharpest of the two or more color channels;
- generating a depth map of the scene based on the auto-focus sweep;
- and
- identifying at least one of the objects as a foreground object or a background object, or one or more of each, based on the determining of the different distances.
Dependencies are inferred from the claim text.