Method of automatically identifying and modifying the appearance of an object in successive frames of a video sequence

• US 6,704,045 B1
• Filed: 09/12/1997
• Issued: 03/09/2004
• Est. Priority Date: 09/12/1996
• Status: Expired due to Fees

• Current Assignee: Pandora International Limited
• Original Assignee: Pandora International Limited
• Inventors: Brett, Stephen
• Primary Examiner(s): Ho, Tuan
• Application Number:
  US08/928,282
• Time in Patent Office: 2,370 Days
• Field of Search: 348/223, 348/229, 348/256, 348/598, 348/599, 348/97, 348/96, 348/576, 348/577, 348/34, 348/221, 348/700, 348/701, 382/256, 382/162, 382/167, 358/505, 358/506, 358/518, 358/519, 358/520, 358/521
• US Class Current: 348/222.1
• CPC Class Codes: G11B 27/28 by using information signals recorded by the same method as the main recording G11B27/22 takes precedence
• CPC Class Codes: H04N 19/142 Detection of scene cut or scene change
• CPC Class Codes: H04N 19/179 the unit being a scene or a shot
• CPC Class Codes: H04N 19/87 involving scene cut or scene change detection in combination with video compression
• CPC Class Codes: H04N 5/147 Scene change detection
• CPC Class Codes: H04N 5/20 Circuitry for controlling amplitude response
• CPC Class Codes: H04N 5/52 Automatic gain control
• CPC Class Codes: H04N 9/11 Scanning of colour motion picture films, e.g. for telecine

Abstract

The invention provides a method of digitally processing a sequence of video frames, in which an object is selected in a first frame; digital object data corresponding to the shape of the object is stored on a digital data storage mechanism, together with digital modification data corresponding to a desired modification of at least one appearance attribute of the object; the object is identified in a second frame of the sequence by means of the object data; and the appearance attribute of the object in the second frame is modified in accordance with the stored modification data. The invention also provides a method of analysing a sequence of video frames, in which a plurality of corresponding lines are sampled from a plurality of consecutive frames of the sequence; the content of the sampled lines is compared for corresponding lines in different frames; a measure of the difference in the content of the corresponding lines is calculated; a first signal is produced if the measure of difference exceeds a first predetermined threshold; and a second signal is produced, in the absence of the first signal, if the measure of difference exceeds a second predetermined threshold, lower than the first threshold.

Claims

1. A method of digitally processing a sequence of video frames, comprising:

selecting an object in a first frame;

storing, on digital data storage means, digital data corresponding to a shape of the object, together with digital modification data corresponding to a desired modification of at least one appearance attribute of the object;

automatically identifying the object in subsequent frames of the sequence by comparing the stored object data with digital data representing contents of subsequent frames; and

modifying the appearance attribute of the object in subsequent frames, in accordance with the stored modification data,

wherein a plurality of sets of digital object data representing different possible views of the object in a frame when taken from different directions are stored on the digital storage means, and

wherein the object is automatically identified in a subsequent video frame of the sequence by comparing digital image data representing the contents of the subsequent video frame with the stored sets of digital object data.

2. A method as claimed in claim 1, wherein the object is identified in a subsequent frame if a match error between the stored object data and digital data representing an object in the subsequent frame is less than a limiting value which permits identification of the object despite minor changes in its appearance.

3. A method as claimed in claim 2, wherein the limiting value may be altered by an operator.

4. A method as claimed in claim 1, further comprising using an edge detection technique to locate objects in the subsequent frame;

and

comparing data representing the located objects with the stored digital object data to identify the object.

5. A method of analysing a sequence of video frames, comprising:

comparing content of each digital image element of a plurality of corresponding distinct regions of a first and subsequent frame;

calculating a measure of difference in the content of corresponding regions in the first and subsequent frame.

generating a signal for a region if the measure of difference for that region exceeds a predetermined threshold; and

selecting an object to be modified in a first region of the first frame, and using signals generated for the plurality of regions to determine the or each region in which the object is likely to be located in the subsequent frame.

6. A method of automatically controlling the gain level applied to a sequence of frames transferred from cinematographic film to a digital medium, comprising:

sampling the content of each digital image element of a plurality of corresponding lines from a first and subsequent frame of the sequence;

comparing the content of the sampled lines for corresponding lines in different frames;

calculating a measure of the difference in the content of the corresponding lines in the first and subsequent frame;

producing a signal when the measure of difference exceeds a predetermined threshold;

automatically setting the gain level on production of the signal; and

maintaining the set gain level until a subsequent signal is produced.

7. The method of claim 6, further comprising:

allowing an operator to manually override the automatically set gain level.

8. A method of automatically controlling the gain level applied to a sequence of frames transferred from cinematographic film to a digital medium, comprising:

sampling the content of each digital image element of a plurality of corresponding lines from a fast and subsequent frame of the sequence;

comparing the content of the sampled lines for corresponding lines in different frames;

calculating a measure of the difference in the content of the corresponding lines in the first and subsequent frame;

producing a first signal when the measure of difference exceeds a first predetermined threshold;

producing a second signal when the measure of difference exceeds a second predetermined threshold lower than the first predetermined threshold;

automatically setting the gain level on production of a signal selected from the group consisting of the first signal and the second signal; and

maintaining the set gain level until a subsequent signal is produced.

9. The method of claim 8, further comprising:

allowing an operator to manually override the automatically set gain level.