Object recognition for gesture tracking

摘要

The accuracy of object tracking using relatively low power gesture cameras can be improved by adjusting camera settings to ensure a sufficient level of contrast or texture to enable stereo disparity calculations for the representations of the object in the captured images. For general gesture input, for example, a user might use a variety of objects in various orientations, such that conventional face or object recognition processes may not be sufficient. Further, such processes are very resource intensive. By adjusting the camera settings (e.g., exposure, gain, and/or aperture) to ensure an adequate level of contrast, objects of an appropriate size and location can be tracked for purposes such as gesture input. Once such an object is identified, coordinates for a bounding box or other indicator can be transferred to any camera or sensor to be used for the tracking.

主权项

1. A computing device, comprising:
at least one processor; a first camera having a first field of view; a second camera having a second field of view at least partially overlapping the first field of view; and memory including instructions that, upon being executed by the at least one processor, cause the computing device to:
capture a first series of images using the first camera, each image of the first series of images captured with a different exposure setting applied to the first camera;capture a second series of images using the second camera, each image of the second series of images captured with a different exposure setting applied to the second camera;determine a first image, from the first series of images, including a first region of interest around a first representation of an object in the first image, in which a first contrast level, between the first region and the first representation of the object, satisfies a threshold contrast level;determine a second image, from the second series of images, including a second region of interest around a second representation of the object in the second image, in which a second contrast level, between the second region and the second representation of the object, is within a specified range of contrast levels relative to the first contrast level;determine disparity information using the first image and the second image;determine that a size of the object is within a specified range of sizes based at least in part upon the disparity information;determine that a first distance to the object is within a specified distance based at least in part upon the disparity information;determine pixel coordinates of an image region corresponding to the first representation of the object in the first image; andlocate a third representation of the object in at least one subsequently captured image based at least in part upon the pixel coordinates.