Video comparing using fingerprint representations

• US 8,433,175 B2
• Filed: 10/28/2008
• Issued: 04/30/2013
• Est. Priority Date: 10/28/2008
• Status: Expired due to Fees

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Shakya, Subodh, Yang, Karen, Kumar, Ashok K., Satyamurthy, Prasanna
• Primary Examiner(s): HARVEY, DAVID E
• Application Number:
  US12/259,849
• Publication Number:
  US 20100104259A1
• Time in Patent Office: 1,645 Days
• Field of Search: 386248-250, 386/241, 707/741, 382/190, 382/195, 382/203
• US Class Current: 386/241
• CPC Class Codes: H04N 5/913 for scrambling ; for copy protection

Abstract

Disclosed is a method comprising identifying one or more moving features in a first video via a computing platform, representing the one or more moving features as fingerprint representations via the computing platform, and comparing the fingerprint representations with fingerprint representations of a second video to determine a match via the computing platform.

Claims

1. A method comprising:

identifying one or more moving features in a first video via a computing platform;

representing said one or more moving features as fingerprint representations via said computing platform;

parsing said fingerprint representations into individual components that represent respective portions of said one or more moving features;

assigning weights to said individual components based, at least in part, on said portion of said one or more moving features represented by said individual components; and

comparing said weighted individual components of said fingerprint representations with corresponding individual components of fingerprint representations of a second video to determine a match via said computing platform.

2. The method of claim 1, wherein said fingerprint representations comprise an N-character key, and wherein N is an integer.

3. The method of claim 1, further comprising:

modifying individual frames of said first video via a computing platform, said modifying including;

detecting a difference between or among two or more sequential said individual frames of said first video, and

removing background images from said one or more sequential frames based at least in part on said detected difference; and

identifying one or more keyframes among said modified individual frames so that said one or more keyframes respectively represent a sequence of said modified individual frames.

4. The method of claim 3, wherein said identifying said one or more keyframes is based at least in part on properties of said modified individual frames, and wherein said properties comprise one or more of hue, saturation, and/or focus.

5. The method of claim 3, wherein said one or more keyframes include said one or more moving features.

6. The method of claim 5, wherein a single keyframe of said one or more keyframes includes more than one moving feature, and wherein individual moving features among said more than one moving feature are represented by respective fingerprint representations.

7. An apparatus comprising:

a memory; and

one or more processors, said one or more processors being adapted to;

identify one or more moving features in a first video;

represent said one or more moving features as fingerprint representations;

parse said fingerprint representations into individual components that represent respective portions of said one or more moving features;

assign weights to said individual components based, at least in part, on said portion of said one or more moving features represented by said individual components; and

compare said weighted individual components of said fingerprint representations with corresponding individual components of fingerprint representations of a second video stored in said memory to determine a match.

8. The apparatus of claim 7, wherein said one or more keyframes include said one or more moving features.

9. The apparatus of claim 7, wherein said fingerprint representations comprise an N-character key, and wherein N is an integer.

10. The apparatus of claim 7, wherein said computing platform is further adapted to:

modify individual frames of said first video by;

detecting a difference between or among two or more sequential said individual frames of said first video, and

removing background images from said one or more sequential frames based at least in part on said detected difference; and

identify one or more keyframes among said modified individual frames so that said one or more keyframes respectively represent a sequence of said modified individual frames.

11. An article comprising:

a non-transitory storage medium comprising machine-readable instructions stored thereon which, if executed by a computing platform, are adapted to enable said computing platform to;

identify one or more moving features in a first video;

represent said one or more moving features as fingerprint representations;

parse said fingerprint representations into individual components that represent respective portions of said one or more moving features;

assign weights to said individual components based, at least in part, on said portion of said one or more moving features represented by said individual components; and

compare said weighted individual components of said fingerprint representations with corresponding individual components of fingerprint representations of a second video to determine a match.

12. The article of claim 11, wherein said one or more keyframes include said one or more moving features.

13. The article of claim 11, wherein a single keyframe of said one or more keyframes includes more than one moving feature, and wherein individual moving features among said more than one moving feature are represented by respective fingerprint representations.

14. The article of claim 11, wherein said fingerprint representations comprise an N-character key, and wherein N is an integer.

15. The article of claim 11, wherein said machine-readable instructions, if executed by a computing platform, are further adapted to enable said computing platform to:

modify individual frames of said first video by;

detecting a difference between or among two or more sequential said individual frames of said first video, and

removing background images from said one or more sequential frames based at least in part on said detected difference; and

identify one or more keyframes among said modified individual frames so that said one or more keyframes respectively represent a sequence of said modified individual frames.

16. The article of claim 15, wherein said identifying said one or more keyframes is based at least in part on properties of said modified individual frames, and wherein said properties comprise one or more of hue, saturation, and/or focus.