METHOD AND SYSTEM FOR PROJECTOR CALIBRATION

Number: 20180376116
Published: 2018-12-27
Filed: 2018-08-09
Assignee: Disney Enterprises, Inc.
Inventors: Grundhöfer; Anselm
CPC: G01B11/2504; G01B11/2513; G06T3/18; H04N9/3185; H04N9/3194
Verdict: Medium Hardware
Source: Google Patents · FreePatentsOnline

The keeper's note

Projector calibration (PGPUB dup).

Abstract

The present disclosure relates to a method for calibrating a projector. The method includes projecting a test pattern onto a scene or an object within the scene and capturing the test pattern using a camera. Once the test pattern image has been captured by the camera, the method further includes estimating by a processing element a perspective projection matrix, warping the estimated projection matrix based on a non-linear distortion function, and modifying the projector to project light based on the distortion function. The present disclosure also relates to a presentation or projection system including two types of projectors for projecting an output presentation having a second image overlaid on a first image.

Background

TECHNICAL FIELD

The technology described herein relates generally to methods and systems for calibrating one or more projectors.BACKGROUND

Image projectors may be used to project images onto a projection surface, such as a screen or other surface. In some applications, video projectors may be used to enhance, compliment, or otherwise augment objects on the surface to create a dynamic and enjoyable user experience, such as, for example, an amusement park attraction. For example, characters or objects may be projected on a surface that virtually “interact” with real objects on the surface.

Conventional video projectors have a number of limitations. In particular, conventional video projectors have limited color gamut and brightness. Due to these limitations, presentations using only conventional video projectors can appear dull and flat. Further, in situations where ambient lighting is present, the resulting image can appear washed out and unrealistic. On the contrary, laser projectors, such as laser scanning projectors, have increased brightness and color gamut as compared to conventional video projectors. In particular, laser projectors can project pure saturated, i.e. monochromatic red, green, and blue color tones, allowing a significantly wider color gamut than conventional video projectors.

However, known calibration techniques for laser projectors require significant user interaction, are time-intensive, and are often not very accurate. As such, th

Claims

1. A method of calibrating a projector comprising: receiving by a processing element a first calibration image of a first calibration pattern projected onto a projection surface; generating a perspective projection matrix by the processing element by modeling the projector as a pinhole device with a pinhole approximation to provide a first calibration approximation based on the first calibration image and the first calibration pattern; and determining by the processing element a two-dimensional non-linear mapping function to correct inaccuracies in the first calibration approximation resulting from the pinhole approximation by analyzing the first calibration image to compensate for distortion characteristics inherent to the projector. 13. A system for calibrating a laser projector with non-uniform distortion characteristics comprising: a laser projector, wherein the laser projector projects a first calibration pattern onto a three-dimensional projection surface; a camera, wherein the camera captures a first calibration image of the first calibration pattern projected onto the projection surface; and a computing device, in electrical communication with the camera and having at least one processing element, performs the following operations: generates a perspective projection matrix by modeling the laser projector as a pinhole device with a pinhole approximation to provide a first calibration approximation based on the first calibration image; and determines a distortion map for the laser projector that maps two-dimensional positions on a virtual image plane of the laser projector to three-dimensional positions on the three-dimensional projection surface by computing the first calibration approximation to the first calibration image and the first calibration pattern to correct inaccuracies in the first calibration approximation resulting from the pinhole approximation, wherein the distortion map determines distortion inherent to the laser projector. 20. A projection system comprising: a video projector projecting a first image onto a projection surface; and a laser projector projecting a second image onto the projection surface, wherein the laser projector is calibrated using a perspective projection matrix by modeling the laser projector as a pinhole device with a pinhole approximation to provide a first calibration approximation based on the first image and a non-linear mapping function to correct inaccuracies in the first calibration approximation resulting from the pinhole approximation and determine a distortion map to adjust for distortion intrinsic to the laser projector; wherein the second image overlays the first image on the projection surface.