Use of 3D depth map with low and high resolution 2D images for gesture recognition and object tracking systems

Disclosed herein according to one embodiment is an object tracker for tracking an object. The object tracker includes a depth camera for acquiring a 3D depth map and a low-resolution 2D image of the object, and an image camera for acquiring a high-resolution 2D image of the object. A processor is in...

Full description

Saved in:
Bibliographic Details
Main Authors Rossa Brian Frederick, Tremaine Jamie Douglas, Hinds Oliver Pilgrim
Format Patent
LanguageEnglish
Published 06.02.2018
Subjects
CALCULATING
COMPUTING
COUNTING
IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:Disclosed herein according to one embodiment is an object tracker for tracking an object. The object tracker includes a depth camera for acquiring a 3D depth map and a low-resolution 2D image of the object, and an image camera for acquiring a high-resolution 2D image of the object. A processor is in communication with the depth camera and the image camera. The processor is configured to: identify a region of interest within the 3D depth map; identify an approximate object silhouette within the low-resolution 2D image using information about the region of interest from the 3D depth map; identify the approximate object silhouette in the high-resolution 2D image using information about the approximate object silhouette from the low-resolution 2D image; generate a refined object silhouette within the high-resolution 2D image using a region segmentation process; and track the object using the refined object silhouette.
Bibliography:Application Number: US201514963992