Extracting photographic images from video

• US 20020028026A1
• Filed: 08/20/2001
• Published: 03/07/2002
• Est. Priority Date: 06/11/1998
• Status: Abandoned Application

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Brandt, Jonathan Worthen, Chen, Shenchang Eric
• Application Number:
  US09/933,617
• Publication Number:
  US 20020028026A1
• Time in Patent Office: Days
• US Class Current: 382/284
• CPC Class Codes: G06F 16/739 in form of a video summary, e.g. the video summary being a video sequence, a composite still image or having synthesized frames
• CPC Class Codes: G06F 16/786 using motion, e.g. object motion or camera motion
• CPC Class Codes: G06T 13/00 Animation
• CPC Class Codes: G06T 3/4053 based on super-resolution, i.e. the output image resolution being higher than the sensor resolution
• CPC Class Codes: H04N 1/40 Picture signal circuits H04N1/387 takes precedence
• CPC Class Codes: H04N 5/147 Scene change detection
• CPC Class Codes: H04N 5/4448 for frame-grabbing
• CPC Class Codes: H04N 5/781 on disks or drums
• CPC Class Codes: H04N 5/782 on tape
• CPC Class Codes: H04N 5/85 on discs or drums

Abstract

Generating photographs from a video. Segments of the video for which frame-to-frame background motion is less than a threshold are identified. For each of the segments, video frames in the segment are combined to generate a photograph representative of the segment.

Claims

1. A method of generating photographs from a video, the method comprising:

identifying segments of the video for which frame-to-frame background motion is less than a threshold; and

for each of the segments, combining video frames in the segment to generate a photograph representative of the segment.

2. The method of claim 1 further comprising:

automatically detecting a scene cut in the video; and

selecting at least one video frame of a segment of the video that follows the scene cut to be a photograph.

3. The method of claim 1 wherein combining the video frames to generate a photograph comprises stitching images in the video frames together to generate a panoramic photograph.

4. The method of claim 1 wherein combining video frames to generate a photograph comprises blending pixels from the video frames to generate a photograph having higher resolution than any one of the video frames.

5. The method of claim 1 wherein combining video frames to generate a photograph comprises blending pixels from the video frames to form a multi-resolution photograph.

6. The method of claim 1 wherein identifying segments of the video for which frame-to-frame background motion is less than a threshold comprises identifying a succession of frames of the video that each include a portion of an image in a preceding frame.

7. The method of claim 6 wherein identifying the succession of frames of the video that each include a portion of an image in a preceding frame comprises removing a dynamic object from at least one frame of the succession of frames before comparing the at least one frame to a preceding frame in the succession of frames.

8. A method comprising:

receiving a video from a customer on a machine-readable medium;

processing the video to generate a set of photographs in return for a fee.

9. The method of claim 8 further comprising recording the set of photographs on the machine-readable medium and returning the machine-readable medium to the customer.

10. The method of claim 8 wherein receiving a video from a customer on a machine-readable medium comprises receiving the video in a data signal propagated over a communications network.

11. The method of claim 8 wherein receiving a video from a customer on a machine-readable comprises receiving the video on a machine-readable diskette.

12. The method of claim 8 wherein processing the video to generate a set of photographic images comprises:

identifying segments of the video that exhibit background motion less than a threshold; and

combining video frames in each of the segments of the video to form the set of photographic images.

13. The method of claim 12 wherein combining video frames in each of the segments of the video to form the set of photographic images comprises stitching together images in the video frames of at least one of the segments of the video to form a panoramic photograph.

14. The method of claim 12 wherein combining video frames in each of the segments of the video to form the set of photographic images comprises stitching together images in the video frames of at least one of the segments of the video to form a photograph having higher pixel resolution than any one of the video frames.

15. The method of claim 8 further comprising posting the set of photographic images on a server that is accessible to the customer via a computer network.

16. The method of claim 8 wherein processing the video to generate a set of photographs comprises printing the set of photographs.

20. The article of claim 19 wherein the one or more computer-readable media comprises a portable storage medium in which at least a portion of the program of instructions is embodied.

21. The article of claim 19 wherein the one or more computer-readable media comprises a propagated data signal in which the program of instructions is embodied.

17. An apparatus for generating photographs from a video, the apparatus comprising:

a scene change estimator to identify segments of the video for which frame-to-frame background motion is less than a threshold; and

a still image constructor to combine video frames in the segment to generate a photograph representative of the segment.

18. An apparatus for generating photographs from a video, the apparatus comprising:

means for identifying segments of the video for which frame-to-frame background motion is less than a threshold; and

means for combining video frames in the segment to generate a photograph representative of the segment.

19. An article of manufacture including one or more computer-readable media that embody a program of instructions for generating photographs from a video, wherein the program of instructions, when executed by a processing unit, causes the processing unit to:

identify segments of the video for which frame-to-frame background motion is less than a threshold; and

for each of the segments, combine video frames in the segment to generate a photograph representative of the segment.