Visual language classification system

• US 7,606,397 B2
• Filed: 12/07/2000
• Issued: 10/20/2009
• Est. Priority Date: 12/14/1999
• Status: Expired due to Fees

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Kowald, Julie Rae
• Primary Examiner(s): Bella; Matthew C.
• Assistant Examiner(s): PERUNGAVOOR, SATHYANARAYA V
• Application Number:
  US09/730,573
• Publication Number:
  US 20030002715A1
• Time in Patent Office: 3,239 Days
• Field of Search: 382/115, 382/117, 382/118, 382/199, 382/209, 382/224, 382/263, 382/264, 382/282, 382/309, 348/77, 348/169
• US Class Current: 382/118
• CPC Class Codes: G06V 40/16 Human faces, e.g. facial parts, sketches or expressions
• CPC Class Codes: G11B 27/034 on discs G11B27/036, G11B27/038 take precedence
• CPC Class Codes: G11B 27/11 by using information not detectable on the record carrier
• CPC Class Codes: G11B 27/28 by using information signals recorded by the same method as the main recording G11B27/22 takes precedence

Abstract

Disclosed are a method and system for automated classification of a digital image. The method analyzes the image for the presence of a human face. A determination is then made regarding the size of the located face compared to the size of the image to classify the image based on the relative size of the face. Alternatively, the position of the face within the image can be used to determine the classification. With a classified image, particularly forming part of a sequence of classified images, editing of the sequence may be performed dependent upon the classification to achieve a desired aesthetic effect. The editing may be performed with the aid of an editing template.

Claims

1. A method of processing an input sequence of digital images, said method comprising:

using a computer to classify each digital image of the input sequence, wherein said classifying comprises;

analyzing the digital image to detect if a human face is present;

determining a relative size of the detected face with respect to a size of the image;

classifying the digital image by shot type based on the relative size of the detected face with respect to the image; and

replaying a part of the input sequence corresponding to the digital image having a shot type of close-up at a slower speed than other shot types.

2. A method according to claim 1, said method further comprising, replaying the input sequence at a faster speed if no human face is detected.

3. A method according to claim 1, said method further comprising:

storing the classification of the digital image as metadata associated with the digital image;

establishing an editing template for the sequence according to the metadata, the template having editing rules corresponding to the method; and

editing the input sequence according to the template based on the metadata.

4. A method according to claim 3, wherein said editing step comprises applying at least one editing rule of the template to each image of the input sequence, wherein the images satisfying the editing rule are included in the output sequence.

5. A method according to claim 4, wherein the editing rule comprises at least one effect for application to the image, the effect being selected from the group consisting of visual effects and audible effects.

6. A method according to claim 5, wherein the visual effects are selected from the group consisting of reproduction speed variation, zooming, blurring, and color variation.

7. A method according to claim 1, wherein said other shot types include at least a medium shot type, a medium long shot type, and a long shot type.

8. An apparatus for processing an input sequence of digital images, said apparatus comprising:

means for classifying each digital image of the input sequence, wherein said means for classifying comprises;

means for analyzing the digital image to detect if a human face is present;

means for determining a relative size of the detected face with respect to a size of the image; and

means for classifying the digital image by shot type based on the relative size of the detected face with respect to the image; and

means for replaying a part of the input sequence corresponding to the digital image having a shot type of close-up at a slower speed than other shot types.

9. A computer-readable storage medium storing a computer-executable program, the computer-executable program being executable by a computer apparatus so as to control the computer apparatus to process an input sequence of digital images, said program comprising:

code for classifying each digital image of the sequence, wherein said code for classifying comprises;

code for analyzing the digital image to detect if a human face is present;

code for determining a relative size of the detected face with respect to a size of the image;

code for classifying the digital image by shot type based on the relative size of the face with respect to the image; and

code for replaying a part of the input sequence corresponding to the digital image having a shot type of close-up at a slower speed than other shot types.