Signal recording/reproducing apparatus, signal recording/reproducing method, signal reproducing apparatus, signal reproducing method, and program

• US 20040218896A1
• Filed: 06/14/2004
• Published: 11/04/2004
• Est. Priority Date: 05/20/2002
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Nishiguchi, Masayuki, Abe, Mototsugu
• Granted Patent: US 7,502,547 B2
• Time in Patent Office: Days
• US Class Current: 386/46
• CPC Class Codes: G11B 2220/2562 DVDs [digital versatile discs]; Digital video discs; MMCDs; HDCDs
• CPC Class Codes: G11B 2220/90 Tape-like record carriers
• CPC Class Codes: G11B 27/105 of operating discs
• CPC Class Codes: G11B 27/107 of operating tapes
• 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: G11B 27/328 on a tape [TTOC]
• CPC Class Codes: G11B 27/329 on a disc [VTOC]
• CPC Class Codes: G11B 27/34 Indicating arrangements indicating means incorporated in magazine or cassette G11B23/046 and G11B23/0875; indicating measured values in general G01D
• CPC Class Codes: G11B 27/36 Monitoring, i.e. supervising the progress of recording or reproducing for digital recording G11B20/00 and s.gr.; for monitoring, testing or measuring of TV recorders of the type covered by H04N5/76 and subgroups, see H04N17/06
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

This invention is a signal recording/reproducing apparatus (1) for searching for a reproduction position desired by a user from a recorded video signal and/or audio signal and reproducing the signal. The apparatus has a feature vector generating unit (30). When recording a video signal/audio signal, the feature vector generating unit (30) generates a feature vector characterizing the video signal and/or audio signal and also records this feature vector to a recording unit (10). When the recorded video signal/audio signal is reproduced, a registration unit (40) registers feature vectors near a position designated by the user, as bookmarks. When the user designates one of the bookmarks, a coincidence detecting unit (60) searches for a coincident position on the basis of the feature vector of the bookmark and the recorded feature vector, and a reproducing unit (20) starts reproduction of the video signal and/or audio signal at a reproduction position set by a reproduction point setting unit (70) on the basis of the coincident position.

Claims

1. A signal recording/reproducing apparatus comprising:

feature vector generation means for extracting a feature quantity from a video signal and/or an audio signal and forming its vector to generate a feature vector;

recording means for recording the video signal and/or the audio signal and the feature vector generated by the feature vector generation means;

registration means for registering at least a part of the feature vectors recorded in the recording means;

coincidence detection means for detecting a feature vector substantially coincident with at least one of feature vectors registered by the registration means, from the feature vector recorded in the recording means;

reproduction point setting means for setting a reproduction start point based on the detected feature vector; and

reproduction means for reproducing the video signal and/or the audio signal from the reproduction start point set by the reproduction point setting means.

2. The signal recording/reproducing apparatus as claimed in claim 1, wherein the feature vector is a time series of a vector formed from a pixel value of an average image of at least one frame within a predetermined time interval of the video signal, or a time series of a vector formed from a pixel value of a small image formed by dividing the average image by predetermined block.

3. The signal recording/reproducing apparatus as claimed in claim 1, wherein the feature vector is a time series of a vector formed from a short-time power spectrum of the audio signal.

4. The signal recording/reproducing apparatus as claimed in claim 1, wherein when reproducing the video signal and/or the audio signal recorded in the recording means, the registration means registers the feature vector corresponding to a signal position designated by a user.

5. The signal recording/reproducing apparatus as claimed in claim 1, wherein when reproducing the video signal and/or the audio signal recorded in the recording means, the registration means registers the feature vector corresponding to a signal position preceding a signal position designated by a user by a predetermined period of time.

6. The signal recording/reproducing apparatus as claimed in claim 1, wherein the coincidence detection means detects a feature vector having a correlation value exceeding a predetermined threshold value with respect to at least one of the registered feature vectors, from the feature vectors recorded in the recording means.

7. The signal recording/reproducing apparatus as claimed in claim 1, further comprising hierarchical conversion means for performing hierarchical conversion of the feature vector and generating a converted vector,

wherein the recording means records the video signal and/or the audio signal and the converted vector generated by the hierarchical conversion means, and

the registration means registers at least a part of the converted vectors recorded in the recording means.

8. The signal recording/reproducing apparatus as claimed in claim 7, wherein the hierarchical conversion means performs hierarchical conversion that repeats processing to divide components of the feature vector into sum components and difference components with respect to adjacent components and then further divide the sum components into sum components and difference components with respect to adjacent components, thus generating the converted vector.

9. The signal recording/reproducing apparatus as claimed in claim 8, wherein the coincidence detection means performs hierarchical calculation of correlation between at least one of the registered converted vectors and the converted vector recorded in the recording means, and detects a feature vector having a correlation value exceeding a predetermined threshold value with respect to at least one of the registered converted vectors, from the converted vector recorded in the recording means.

10. The signal recording/reproducing apparatus as claimed in claim 9, wherein the coincidence detection means performs hierarchical calculation of correlation to calculation correlation by increasing the number of components so as to sequentially include difference components from sum components of an upper hierarchical level of the converted vector, and if a correlation value on an intermediate hierarchical level in the hierarchical calculation of correlation is lower than the threshold value, the coincidence detection means does not perform further calculation of correlation.

11. A signal recording/reproducing method comprising:

a feature vector generation step of extracting a feature quantity from a video signal and/or an audio signal and forming its vector to generate a feature vector;

a recording step of recording the video signal and/or the audio signal and the feature vector generated at the feature vector generation step into recording means;

a registration step of registering at least a part of the feature vectors recorded in the recording means;

a coincidence detection step of detecting a feature vector substantially coincident with at least one of feature vectors registered at the registration step, from the feature vector recorded in the recording means;

a reproduction point setting step of setting a reproduction start point based on the detected feature vector; and

a reproduction step of reproducing the video signal and/or the audio signal from the reproduction start point set at the reproduction point setting step.

12. The signal recording/reproducing method as claimed in claim 11, wherein at the registration step, when reproducing the video signal and/or the audio signal recorded in the recording means, the feature vector corresponding to a signal position designated by a user is registered.

13. The signal recording/reproducing method as claimed in claim 11, wherein at the coincidence detection step, a feature vector having a correlation value exceeding a predetermined threshold value with respect to at least one of the registered feature vectors is detected from the feature vectors recorded in the recording means.

14. The signal recording/reproducing method as claimed in claim 11, further comprising a hierarchical conversion step performing hierarchical conversion of the feature vector and generating a converted vector,

wherein at the recording step, the video signal and/or the audio signal and the converted vector generated by the hierarchical conversion means are recorded into the recording means, and

at the registration step, at least a part of the converted vectors recorded in the recording means is registered.

15. The signal recording/reproducing method as claimed in claim 14, wherein at the hierarchical conversion step, hierarchical conversion that repeats processing to divide components of the feature vector into sum components and difference components with respect to adjacent components and then further divide the sum components into sum components and difference components with respect to adjacent components is performed, thus generating the converted vector.

16. The signal recording/reproducing method as claimed in claim 15, wherein at the coincidence detection step, hierarchical calculation of correlation between at least one of the registered converted vectors and the converted vector recorded in the recording means is performed, and a feature vector having a correlation value exceeding a predetermined threshold value with respect to at least one of the registered converted vectors is detected from the converted vector recorded in the recording means.

17. The signal recording/reproducing method as claimed in claim 16, wherein at the coincidence detection step, hierarchical calculation of correlation to calculation correlation by increasing the number of components so as to sequentially include difference components from sum components of an upper hierarchical level of the converted vector is performed, and if a correlation value on an intermediate hierarchical level in the hierarchical calculation of correlation is lower than the threshold value, no further calculation of correlation is performed.

18. A program for causing a computer to execute predetermined processing, the program comprising:

a feature vector generation step of extracting a feature quantity from a video signal and/or an audio signal and forming its vector to generate a feature vector;

a recording step of recording the video signal and/or the audio signal and the feature vector generated at the feature vector generation step into recording means;

a registration step of registering at least a part of the feature vectors recorded in the recording means;

a coincidence detection step of detecting a feature vector substantially coincident with at least one of feature vectors registered at the registration step, from the feature vector recorded in the recording means;

a reproduction point setting step of setting a reproduction start point based on the detected feature vector; and

a reproduction step of reproducing the video signal and/or the audio signal from the reproduction start point set at the reproduction point setting step.

19. The program as claimed in claim 18, wherein at the registration step, when reproducing the video signal and/or the audio signal recorded in the recording means, the feature vector corresponding to a signal position designated by a user is registered.

20. The program as claimed in claim 18, wherein at the coincidence detection step, a feature vector having a correlation value exceeding a predetermined threshold value with respect to at least one of the registered feature vectors is detected from the feature vectors recorded in the recording means.

21. The program as claimed in claim 18, further comprising a hierarchical conversion step performing hierarchical conversion of the feature vector and generating a converted vector,

wherein at the recording step, the video signal and/or the audio signal and the converted vector generated by the hierarchical conversion means are recorded into the recording means, and

at the registration step, at least a part of the converted vectors recorded in the recording means is registered.

22. The program as claimed in claim 21, wherein at the hierarchical conversion step, hierarchical conversion that repeats processing to divide components of the feature vector into sum components and difference components with respect to adjacent components and then further divide the sum components into sum components and difference components with respect to adjacent components is performed, thus generating the converted vector.

23. The program as claimed in claim 22, wherein at the coincidence detection step, hierarchical calculation of correlation between at least one of the registered converted vectors and the converted vector recorded in the recording means is performed, and a feature vector having a correlation value exceeding a predetermined threshold value with respect to at least one of the registered converted vectors is detected from the converted vector recorded in the recording means.

24. The program as claimed in claim 23, wherein at the coincidence detection step, hierarchical calculation of correlation to calculation correlation by increasing the number of components so as to sequentially include difference components from sum components of an upper hierarchical level of the converted vector is performed, and if a correlation value on an intermediate hierarchical level in the hierarchical calculation of correlation is lower than the threshold value, no further calculation of correlation is performed.

25. A recording medium on which a program for causing a computer to execute predetermined processing is recorded, the program comprising:

a feature vector generation step of extracting a feature quantity from a video signal and/or an audio signal and forming its vector to generate a feature vector;

a recording step of recording the video signal and/or the audio signal and the feature vector generated at the feature vector generation step into recording means;

a registration step of registering at least a part of the feature vectors recorded in the recording means;

a coincidence detection step of detecting a feature vector substantially coincident with at least one of feature vectors registered at the registration step, from the feature vector recorded in the recording means;

a reproduction point setting step of setting a reproduction start point based on the detected feature vector; and

a reproduction step of reproducing the video signal and/or the audio signal from the reproduction start point set at the reproduction point setting step.

26. The recording medium as claimed in claim 25, wherein at the registration step, when reproducing the video signal and/or the audio signal recorded in the recording means, the feature vector corresponding to a signal position designated by a user is registered.

27. The recording medium as claimed in claim 25, wherein at the coincidence detection step, a feature vector having a correlation value exceeding a predetermined threshold value with respect to at least one of the registered feature vectors is detected from the feature vectors recorded in the recording means.

28. The recording medium as claimed in claim 25, further comprising a hierarchical conversion step performing hierarchical conversion of the feature vector and generating a converted vector,

wherein at the recording step, the video signal and/or the audio signal and the converted vector generated by the hierarchical conversion means are recorded into the recording means, and

at the registration step, at least a part of the converted vectors recorded in the recording means is registered.

29. The recording medium as claimed in claim 28, wherein at the hierarchical conversion step, hierarchical conversion that repeats processing to divide components of the feature vector into sum components and difference components with respect to adjacent components and then further divide the sum components into sum components and difference components with respect to adjacent components is performed, thus generating the converted vector.

30. The recording medium as claimed in claim 29, wherein at the coincidence detection step, hierarchical calculation of correlation between at least one of the registered converted vectors and the converted vector recorded in the recording means is performed, and a feature vector having a correlation value exceeding a predetermined threshold value with respect to at least one of the registered converted vectors is detected from the converted vector recorded in the recording means.

31. The recording medium as claimed in claim 30, wherein at the coincidence detection step, hierarchical calculation of correlation to calculation correlation by increasing the number of components so as to sequentially include difference components from sum components of an upper hierarchical level of the converted vector is performed, and if a correlation value on an intermediate hierarchical level in the hierarchical calculation of correlation is lower than the threshold value, no further calculation of correlation is performed.

32. A signal reproducing apparatus for reproducing a video signal and/or an audio signal recorded on a recording medium, the apparatus comprising:

designation means for designating a reproduction start position of the video signal and/or the audio signal;

reproduction point setting means for setting a position temporally preceding the position designated by the designation means, as a reproduction start point; and

reproduction means for starting reproduction of the video signal and/or the audio signal at the reproduction start point set by the reproduction point setting means.

33. The signal reproducing apparatus as claimed in claim 32, further comprising commercial message detection means for detecting a commercial message in the video signal and/or in the audio signal,

wherein the reproduction point setting means sets an end position of a commercial message temporally preceding the position designated by the designation means, as a reproduction start point.

34. The signal reproducing apparatus as claimed in claim 32, further comprising time detection means for detecting the time when the video signal and/or the audio signal is recorded to the recording medium,

wherein the reproduction point setting means sets a predetermined time point temporally preceding the position designated by the designation means, as a reproduction start point.

35. The signal reproducing apparatus as claimed in claim 34, wherein the predetermined time point is a start time of a program including the position designated by the designation means.

36. The signal reproducing apparatus as claimed in claim 34, wherein the predetermined time is the 0^th minute of every hour, the 15^th minute of every hour or the 30^th minute of every hour.

37. A signal reproducing method for reproducing a video signal and/or an audio signal recorded on a recording medium, the method comprising:

a designation step of designating a reproduction start position of the video signal and/or the audio signal;

a reproduction point setting step of setting a position temporally preceding the position designated at the designation step, as a reproduction start point; and

a reproduction step of starting reproduction of the video signal and/or the audio signal at the reproduction start point set at the reproduction point setting step.

38. The signal reproducing method as claimed in claim 37, further comprising a commercial message detection step of detecting a commercial message in the video signal and/or in the audio signal,

wherein at the reproduction point setting step, an end position of a commercial message temporally preceding the position designated at the designation step is set as a reproduction start point.

39. The signal reproducing method as claimed in claim 37, further comprising a time detection step of detecting the time when the video signal and/or the audio signal is recorded to the recording medium,

wherein at the reproduction point setting step, a predetermined time point temporally preceding the position designated at the designation step is set as a reproduction start point.

40. The signal reproducing method as claimed in claim 39, wherein the predetermined time point is a start time of a program including the position designated at the designation step.

41. The signal reproducing method as claimed in claim 39, wherein the predetermined time is the 0^th minute of every hour, the 15^th minute of every hour or the 30^th minute of every hour.