System and method for tracking objects with projected m-sequences

Number: 10126123
Published: 2018-11-13
Filed: 2015-09-21
Assignee: CARNEGIE MELLON UNIVERSITY; DISNEY ENTERPRISES, INC.
Inventors: Harrison; Christopher; Xiao; Bo Robert; Hudson; Scott E.; Poupyrev; Ivan; Willis; Karl D. D.
CPC: G06F3/0346; G01S5/16; G06F3/0317; G06F3/0304; G01C3/00
Verdict: Low Notable software
Source: Google Patents · FreePatentsOnline

The keeper's note

Structured-light m-sequence object tracking.

Abstract

According to embodiments of the present invention are a system and method that use projected structured patterns of light and linear optical sensors for motion tracking. Sensors are capable of recovering two-dimensional location within the projection area, while several sensors can be combined for up to six degrees of freedom tracking. The structure patterns are based on m-sequences, in which any consecutive subsequence of m bits is unique. Both digital and static light sources can be used. The system and method of the present invention enables high-speed, high precision, and low-cost motion tracking for a wide range of applications.

Background

BACKGROUND OF THE INVENTION(1) The invention relates generally to the use of structured patterns of light for motion tracking of objects. More specifically, the invention relates to a system and method where projected structured patterns of light based on m-sequences are used in combination with an optical sensor to locate and track an object.(2) Motion tracking is a fundamental component of almost all interactive systems. The rapidly growing area of embodied and gestural interaction underlines the continuing importance of accurate, fast, and affordable motion tracking technology. Numerous approaches have been developed to sense and track motion using mechanical, inertial, acoustic, magnetic, and radio-based sensing techniques.(3) In one example of an existing tracking system, natural features in a scene can be tracked using standard cameras to recover three-dimensional information from a series of two-dimensional images, for example, by using Structure from Motion (SfM). When combined with Simultaneous Localization and Mapping (SLAM), camera pose can be estimated within a dynamically constructed map of the environment. Although powerful, natural feature tracking techniques are computationally expensive and challenging to implement in embedded systems and mobile devices.(4) In another example, marker-based motion tracking systems instrument objects and the environment with physical markers to aid the tracking process. A camera is used to detect the markers in the scene and es

Claims

1. A method of tracking an object affixed to one of a projector and an optical sensor, comprising: generating an m-sequence comprised of a series of subsequences; projecting a structured light pattern from a projector, forming a projected field, wherein the structured light pattern is based on the m-sequence; positioning an optical sensor within the projected field to receive a portion of the structured light pattern, wherein the portion of the structured light pattern received by the sensor contains at least one subsequence of the series of subsequences; generating data with the optical sensor based on the received subsequence; thresholding the data from the optical sensor; transforming the thresholded data into a binary string, wherein transforming the thresholded data into a binary string comprises testing an offset and step size of the thresholded data to transform the thresholded data into a series of bits; determining if the binary string exists in the m-sequence; and calculating a position of the object based on a location of the received subsequence within the structured light pattern. 13. A system for tracking an object using a structured light pattern, the system comprising: a projector for projecting a structured light pattern, the projector comprising: a light source for emitting light, and an image source for modulating the emitted light into the structured light pattern, wherein the structured light pattern is based on an m-sequence; a sensor located in an area illuminated by the structured light pattern, wherein the sensor generates a data signal based on a portion of the structured light pattern irradiating the sensor; and a processor in communication with the sensor, wherein the processor is configured to process the data signal received from the sensor to determine a location of the sensor based on the portion of the structured light pattern irradiating the sensor and the m-sequence, wherein the data signal is thresholded and transformed into a binary string by testing an offset and step size of the thresholded data signal to transform the thresholded data into a series of bits.