Method and apparatus for unencumbered capture of an object

• US 6,101,289 A
• Filed: 10/15/1997
• Issued: 08/08/2000
• Est. Priority Date: 10/15/1997
• Status: Expired due to Term

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Planet Electric Incorporated
• Inventors: Kellner, Charles R.
• Primary Examiner(s): COUSO, YON JUNG
• Application Number:
  US08/951,087
• Time in Patent Office: 1,028 Days
• Field of Search: 382/284, 382/294, 382/275, 382/174, 382/100, 382/276, 348/586, 348/590, 348/598, 348/170, 348/587, 345/115, 345/121
• US Class Current: 382/276
• CPC Class Codes: G06T 2207/20008 Globally adaptive
• CPC Class Codes: G06T 7/194 involving foreground-background segmentation
• CPC Class Codes: H04N 5/272 Means for inserting a foreground image in a background image, i.e. inlay, outlay

Abstract

A system and method are disclosed for providing an unencumbered capture by a computer system of an object in real-time. An object is placed in a displayer having an appropriate blue-screen surface which is positioned in the view of a camera connected to a computer. The camera captures a frame of an input image which included the object within the displayer being held by a user and possibly other background items. The system identifies a reduced-noise image of the input image and begins isolating the object. Any background items are filled out thereby recognizing the existence of the displayer. The displayer is then filled out leaving the object to be captured and entered into the computer for further manipulation and augmentation. The process is done with minimal human intervention and relies entirely on real-time visual input.

Claims

1. A computer-implemented method of capturing an image of an object comprising:

obtaining a frame of an input image that contains an object to be captured;

determining an input image noise level wherein the input image noise level is ascertained by examining properties of the input image and a reference image;

determining whether the input image noise level is preferable to a noise level of an aged reduced-noise image derived from a previously obtained input image;

isolating an object image of the object in the input image such that areas surrounding the object are determined to be separate from the object; and

inputting the object image into a computer thereby allowing the object image to be digitally manipulated.

2. A method as recited in claim 1 wherein obtaining a frame of an input image further includes stabilizing the object in a displayer made of an appropriate blue-screen material.

3. A method as recited in claim 1 wherein obtaining a frame of an input image further includes displaying the object in front of a camera.

4. A method as recited in claim 1 wherein determining the input image noise level further includes creating the reference image from pixels whose properties are derived from properties of pixels in the input image.

5. A method as recited in claim 1 wherein determining the input image noise level further includes maintaining a running average of visual differences between the input image and the reference image.

6. A method as recited in claim 1 wherein the aged reduced-noise image has a noise level that is adjusted when a new input image is obtained.

7. A method as recited in claim 1 wherein determining whether the input image noise level is preferable to a noise level of an aged reduced-noise image further includes comparing the input image noise level to a threshold noise level.

8. A method as recited in claim 1 wherein isolating an object image further includes determining whether a displayer surrounds the object being captured.

9. A method as recited in claim 1 wherein isolating an input image further includes creating one or more histograms for an image derived from the input image.

10. A method as recited in claim 1 wherein isolating an image of the object further includes isolating a displayer in the input image.

11. A computer readable medium including program instructions implementing the process of claim 1.

12. A computer-implemented system for capturing an image of an object, the system comprising:

a frame capturer for obtaining a frame of an input image that contains an object to be captured;

a noise-level detector for determining the noise level of an input image wherein the noise level is ascertained by examining properties of the input image and a reference image;

a noise-level comparator for determining whether the input image noise level is preferable to a noise level aged reduced-noise image derived from a previously obtained input image;

an image isolator for isolating an object image of the object and input image such that areas surrounding the object are determined to be separate from the object; and

an image transmitter for inputting the image into a computer thereby allowing the object image to be digitally manipulated.

13. A system as recited in claim 12 wherein the frame capture further comprises an object displayer made of an appropriate blue-screen material for stabilizing the object.

14. A system as recited in claim 12 wherein the object is displayed in front of a camera.

15. A system as recited in claim 12 wherein the noise level detector further includes an image creator for creating the reference image from pixels whose properties are derived from properties of pixels in the input image.

16. A system as recited in claim 12 wherein the noise-level detector further comprises a visual comparator for maintaining a running average of visual differences between the input image and the reference image.

17. A system as recited in claim 12 wherein the noise-level comparator further comprises a noise-level adjuster for adjusting a noise level for the aged reduced-noise image when a new input image is obtained.

18. A system as recited in claim 12 wherein the noise-level comparator compares the input image noise-level to a threshold noise level.

19. A system as recited in claim 12 wherein the image isolator further comprises a display dector for determining whether a displayer surrounds the object being captured.

20. A system as recited in claim 12 wherein the image isolator further comprises a histogram creator for creating one or more histograms for an image derived from the input image.

21. A system as recited in claim 12 wherein the image isolator isolates a displayer in the input image.