Determining camera activity from a steadiness signal

• US 8,180,209 B2
• Filed: 05/19/2010
• Issued: 05/15/2012
• Est. Priority Date: 05/19/2010
• Status: Active Grant

• Current Assignee: Apple Inc.
• Original Assignee: Eastman Kodak Company
• Inventors: Gallagher, Andrew C., Deever, Aaron T.
• Primary Examiner(s): LaBalle, Clayton E
• Assistant Examiner(s): Fenwick, Warren
• Application Number:
  US12/782,960
• Publication Number:
  US 20110286731A1
• Time in Patent Office: 727 Days
• Field of Search: 396/53, 396/54, 396/55, 396/52, 382/122, 382/123, 348/208.4, 348/208.14
• US Class Current: 396/53
• CPC Class Codes: G03B 17/00 Details of cameras or camera bodies; Accessories therefor lens hoods or caps G03B11/04

Abstract

A method of identifying the activity of a camera used for capturing an image or video, such activity not being related to the activity of subjects of the scene being captured includes determining a steadiness signal related to movement of the camera while capturing an image or video, wherein the steadiness signal identifies the motion of the camera in at least two dimensions as a function of time; and using a processor for determining a camera activity classification for the camera based on an analysis of the steadiness signal, wherein the camera activity classification includes categories of the camera being stationary, the camera being located in a moving vehicle, the camera being held by hand, or the camera being held by a jogging or walking user.

Claims

1. A method of identifying an activity of a camera used for capturing an image or video, such activity not being related to activities of subjects of a scene being captured, comprising:

(a) determining a steadiness signal related to movement of the camera while capturing an image or video, wherein the steadiness signal identifies the motion of the camera in at least two dimensions as a function of time further including providing at least two frames from a video, and storing the frames in memory and comparing corresponding pixel values of the frames for determining the steadiness signal, and

(b) using a processor for determining a camera activity classification for the camera based on an analysis of the steadiness signal, wherein the motion of the camera activity classification includes categories of the camera being stationary, the camera being located in a moving vehicle, the camera being held by hand, or the camera being held by a jogging or walking user.

2. The method of claim 1, further including an accelerometer integral with the camera responsive to camera movement for determining the steadiness signal and for applying such signal to the processor.

3. The method of claim 1, wherein the compared pixel values are from two adjacent frames.

4. The method of claim 1, further including providing a single image and using pixels from the single image to perform a deconvolution operation to determine the steadiness signal.