Method to generate virtual display surfaces from video imagery of road based scenery
First Claim
1. A method, comprising:
- capturing an image by an image capturing device in a vehicle;
detecting visual features in the image indicative of a position of a road relative of the vehicle by at least;
cropping the captured image to extract a portion of image including outline of a road,performing inverse perspective mapping on the cropped image to generate a bird'"'"'s eye view image,using a separable filter on the bird'"'"'s eye view image to generate a filtered image,detecting line features representing edges of the road from the filtered image, andfitting curves onto the detected line features;
constructing a virtual model of environment in front of the vehicle based on the detected visual features, the virtual model including at least one surface;
generating graphical elements to convey information on a display device in the vehicle;
configuring the generated graphical elements with reference to the at least one surface of the virtual model; and
displaying the configured graphical elements on the display device.
1 Assignment
0 Petitions
Accused Products
Abstract
Generating a virtual model of environment in front of a vehicle based on images captured using an image capturing. The Images captured on an image capturing device of a vehicle are processed to extract features of interest. Based on the extracted features, a virtual model of the environment is constructed. The virtual model includes one or more surfaces. Each of the surfaces may be used as a reference surface to attach and move graphical elements generated to implement augmented reality (AR). As the vehicle moves, the graphical elements move as if the graphical elements are affixed to the one of the surfaces. By presenting the graphical elements to move together with real objects in front of the vehicle, a driver perceives the graphical elements as being part of the actual environment and reduces distraction or confusion associated with the graphical elements.
44 Citations
16 Claims
-
1. A method, comprising:
-
capturing an image by an image capturing device in a vehicle; detecting visual features in the image indicative of a position of a road relative of the vehicle by at least; cropping the captured image to extract a portion of image including outline of a road, performing inverse perspective mapping on the cropped image to generate a bird'"'"'s eye view image, using a separable filter on the bird'"'"'s eye view image to generate a filtered image, detecting line features representing edges of the road from the filtered image, and fitting curves onto the detected line features; constructing a virtual model of environment in front of the vehicle based on the detected visual features, the virtual model including at least one surface; generating graphical elements to convey information on a display device in the vehicle; configuring the generated graphical elements with reference to the at least one surface of the virtual model; and displaying the configured graphical elements on the display device. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. An augmented reality system comprising:
-
an image capturing device installed in a vehicle to capture an image comprising a road in front of the vehicle; an image feature detector coupled to the image capturing device and configured to detect visual features in the image indicative of a position of a road relative to the vehicle by at least; cropping the captured image to extract a portion of image including outline of a road; and performing inverse perspective mapping on the cropped image to generate a bird'"'"'s eye view image, using a separable filtering on the bird'"'"'s eye view image to generate a filtered image, detecting line features representing edges of the road image from the filtered image, and fitting curves onto the detected line features; and a model generator coupled to the image feature detector and configured to construct a virtual model of environment in front of the vehicle based on the visual features, the virtual model including at least one surface; a graphical element generator configured to graphical elements to convey information; a graphical image renderer configured to set configurations of the generated graphical elements with reference to the at least one surface of the virtual model; and a display device coupled to the graphical image renderer and configured to display the generated graphical elements according to the set configurations. - View Dependent Claims (10, 11, 12, 13, 14, 15)
-
-
16. A non-transitory computer-readable storage medium storing instructions thereon, the instructions when executed by a processor causes the processor to:
-
detect visual features in an image indicative of a position of a road relative to the vehicle, the image captured by an image capturing device in the vehicle by at least; cropping the captured image to extract a portion of image including outline of a road, performing inverse perspective mapping on the cropped image to generate a bird'"'"'s eye view image, using a separable filter on the bird'"'"'s eye view image to generate a filtered image, detecting line features representing edges of the road from the filtered image, and fitting curves onto the detected line features; construct a virtual model of environment in front of the vehicle based on the visual features, the virtual model including at least one surface; generate graphical elements to convey information on a display device in the vehicle; configure the generated graphical elements with reference to the at least one surface of the virtual model; and display the configured graphical elements on the display device.
-
Specification