Systems and methods for chronologically ordering digital media and approximating a timeline of an event
Systems and methods synchronize digital audio and/or video tracks recorded at a same event. The systems and methods access input digital audio tracks that are storable in a database, wherein the input digital audio tracks comprise a list of original recorded digital audio tracks from original recorded digital videos that were recorded from multiple digital sources at the same event. The systems and methods extract at least one feature from each digital audio track of the input digital audio tracks, and construct, based on the at least one feature, a peak map for each digital audio track of the input digital audio tracks. The systems and methods detect a match between at least two digital audio tracks by performing a cross-correlation of the peak maps of the digital audio tracks of the input digital audio tracks and/or estimate a delay between at least one pair of digital audio tracks of the input digital audio tracks, wherein the delay comprises a time difference between the at least one pair of digital audio tracks that received a maximum value in the cross-correlation of the peak maps. The systems and method create an output group comprising the at least two digital audio tracks and/or the at least one pair of digital audio tracks.
1. A method for synchronizing at least one track selected from digital audio tracks and digital video tracks recorded at a same event, the method comprising:
accessing input digital audio tracks that are storable in a database, wherein the input digital audio tracks comprise a list of original recorded digital audio tracks from original recorded digital videos that were recorded from multiple digital sources at the same event;
extracting at least one feature from each digital audio track of the input digital audio tracks,
constructing, based on the at least one feature, a peak map for each digital audio track of the input digital audio tracks;
performing a cross-correlation of the peak maps of the digital audio tracks of the input digital audio tracks;
estimating a delay between at least one pair of digital audio tracks of the input digital audio tracks, wherein the delay comprises a time difference between the at least one pair of digital audio tracks that received a maximum value in the cross-correlation of the peak maps; and
creating an output group comprising the at least one pair of digital audio tracks, wherein each digital audio tracks of the output group has an offset inside the output group.
2. The method according to claim 1 , wherein each digital audio tracks of the output group is synchronized such that at least one portion of the same event is isolated by the output group.
3. The method according to claim 1 , further comprising:
providing a database, wherein the input digital audio tracks are stored within the database.
4. The method according to claim 1 , further comprising:
storing the output group within a digital memory device.
5. The method according to claim 1 , wherein the output group is a group of chunks of digital audio and/or video tracks, whereby each chunk has a given offset inside the group of chunks.
6. The method according to claim 1 , wherein each peak map is constructed by searching each digital audio track for a local maxima to construct a binary spectrogram that equals 1 at a local maximum and 0 elsewhere.
7. The method according to claim 6 , further comprising:
determining the local maxima by utilizing a function imdilate, comprising an image and a mask, that determines a maximum value of a region around each pixel defined by the mask.
8. The method according to claim 7 , wherein the mask is an elliptic mask.
9. A method for synchronizing at least one track selected from digital audio tracks and digital video tracks recorded at a same event, the method comprising:
accessing input digital audio tracks that are storable in a database, wherein the input digital audio tracks comprise a list of original recorded digital audio tracks from original recorded digital videos that were recorded from multiple digital sources at the same event;
extracting at least one feature from each digital audio track of the input digital audio tracks,
constructing, based on the at least one feature, a peak map for each digital audio track of the input digital audio tracks;
detecting a match between at least two digital audio tracks by performing a cross-correlation of the peak maps of the digital audio tracks of the input digital audio tracks;
computing a confidence value of the match;
selecting at least one link between the at least two digital audio tracks of the match, wherein the at least one link comprises a confidence value above a predetermined confidence threshold; and
creating an output group based on the at least one link, wherein at least two digital audio tracks belong to the output group if the at least one link exists between the at least two digital audio tracks.
10. The method according to claim 9 , wherein the at least two digital audio tracks of the output group are synchronized such that at least one portion of the same event is isolated by the output group.
11. The method according to claim 9 , further comprising:
providing a database, wherein the input digital audio tracks are stored within the database.
12. The method according to claim 9 , further comprising:
storing the output group within a digital memory device.
13. The method according to claim 9 , further comprising:
utilizing equation (1) to compute the confidence of the match,
confidence= M/M* (1),
wherein M is a maximum value obtained during the cross-correlation and M* is an estimate representing an expected maximum of the cross-correlation if the at least two digital audio tracks of the match were decorrelated and the lengths of the at least two digital audio tracks of the match.
14. The method according to claim 13 , further comprising:
discarding links between at least two digital audio tracks having a confidence value below the predetermined confidence threshold, wherein, optionally, the predetermined confidence threshold is about 6.
15. The method according to claim 9 , further comprising:
detecting a list of links between the digital audio tracks, wherein each link is defined (i) as a delay and a confidence when groups of digital audio tracks are outputted or (ii) a delay, a confidence and match bounds when groups of chunks of the digital audio tracks are outputted.
16. A non-transitory computer-readable medium with instructions stored thereon, that when executed by a microprocessor, perform a method for synchronizing at least one track selected from digital audio tracks and digital video tracks recorded at a same event, the method comprising:
accessing input digital audio tracks that are storable in a database, wherein the input digital audio tracks comprise a list of original recorded digital audio tracks from original recorded digital videos that were recorded from multiple digital sources at the same event;
extracting at least one feature from each digital audio track of the input digital audio tracks,
constructing, based on the at least one feature, a peak map for each digital audio track of the input digital audio tracks;
detecting a match between at least two digital audio tracks by performing a cross-correlation of the peak maps of the digital audio tracks of the input digital audio tracks and/or estimating a delay between at least one pair of digital audio tracks of the input digital audio tracks, wherein the delay comprises a time difference between the at least one pair of digital audio tracks that received a maximum value in the cross-correlation of the peak maps; and
creating an output group comprising the at least two digital audio tracks and/or the at least one pair of digital audio tracks.
17. The non-transitory computer-readable medium according to claim 16 , wherein the at least two digital audio tracks or the at least one pair of digital audio tracks of the output group are synchronized such that at least one portion of the same event is isolated by the output group.
18. The non-transitory computer-readable medium according to claim 16 , further comprising:
providing a database, wherein the input digital audio tracks are stored within the database.
19. The non-transitory computer-readable medium according to claim 16 , further comprising:
storing the output group within a digital memory device.
20. The non-transitory computer-readable medium according to claim 16 , further comprising:
computing, after the match is detected, a confidence value of the match;
selecting at least one link between the at least two digital audio tracks of the match, wherein the at least one link comprises a confidence value above a predetermined confidence threshold; and
creating an output group based on the at least one link, wherein at least two digital audio tracks belong to the output group if the at least one link exists between the at least two digital audio tracks.