Method for estimating volumetric distance maps from 2D depth images

• US 6,262,738 B1
• Filed: 12/04/1998
• Issued: 07/17/2001
• Est. Priority Date: 12/04/1998
• Status: Expired due to Term

• Current Assignee: Mitsubishi Electric Research Laboratories, Inc. (Mitsubishi Electric Corporation)
• Original Assignee: Mitsubishi Electric Research Laboratories, Inc. (Mitsubishi Electric Corporation)
• Inventors: Perry, Ronald N., Gibson, Sarah F. F.
• Primary Examiner(s): Nguyen, Phu K.
• Application Number:
  US09/206,071
• Time in Patent Office: 956 Days
• Field of Search: 345/418, 345/419, 345/420, 345/421, 345/422, 345/424
• US Class Current: 345/419
• CPC Class Codes: G06T 15/06 Ray-tracing
• CPC Class Codes: G06T 15/40 Hidden part removal
• CPC Class Codes: H04N 13/128 Adjusting depth or disparity

Abstract

A volumetric distance map of an object is generated from one or more depth images of the object. Each depth image is projected onto the object by casting parallel rays to the object. The parallel rays are cast perpendicular to the depth image. Sample points in a projected distance volume represent distances from the distance map to a surface of the object. The magnitude of a local gradient at each sample point of the projected distance volume is determined, and each distance at each sample point is divided by the magnitude of the corresponding local gradient at each sample point to obtain a scalar distance to a closest surface of the object.

Claims

1. A method for generating a volumetric distance map of an object, comprising the steps of:

generating a depth image of the object;

projecting the depth image to obtain a projected distance volume of the object, the projected distance volume having a plurality of sample points, each sample point of the projected distance volume representing a distance from the distance map to a surface of the object;

determining a magnitude of a local gradient at each sample point of the projected distance volume;

dividing each distance at each sample point by the magnitude of the corresponding local gradient at each sample point to obtain a scalar distance to a closest surface of the object.

2. The method of claim 1 wherein the distances are determined by casting parallel rays from the depth image to the object, and locating the first point where each ray intersects the object.

3. The method of claim 2 wherein a given sample point at (i, j, k) in the projected distance volume is related to a corresponding point (i, j, 0) in the depth image according to dist(i, j, k)=(i, j, 0)+k, where k is a grid increment perpendicular to the depth image.

4. The method of claim 1 wherein a plurality of depth images is generated.

5. The method of claim 4 wherein the distances of the projected distance volume are interpolated from the plurality of distance maps.

6. The method of claim 1 wherein the volumetric distance map is used to control a robot.

7. The method of claim 1 wherein the volumetric distance map is used to control a haptic device.

8. The method of claim 1 wherein the volumetric distance map is used to shade a rendered image of the object.

9. The method of claim 1 wherein the volumetric distance map is used to generate a model of a real object.

10. The method of claim 1 wherein the depth image is read from a z-buffer of a graphics device.

11. The method of claim 1 wherein the depth image is measured using a scanner.