Multi-channel audio video fingerprinting

A method comprising:nreceiving, by a system including a processor, audio channel data and video channel data associated with a video;generating, by the system, a set of audio fingerprints based on the audio channel data;generating, by the system, a set of mean frames based on a sliding time window and the video channel data;generating, by the system, a set of wavelet min-hashes based on the set of mean frames;generating, by the system, a set of video fingerprints based on the set of wavelet min-hashes, wherein video fingerprints in the set of video fingerprints are min-hashes; andgenerating, by the system, a set of multi-channel fingerprints based on the set of audio fingerprints and the set of video fingerprints.

A system comprising:na processor; anda memory communicatively coupled to the processor, the memory having stored therein computer-executable instructions, comprising:nan input component configured to receive audio channel data and video channel data associated with a video;a spectrogram component configured to generate an audio spectrogram based on the audio channel data;a video processing component configured to generate a set of mean frames based on a sliding time window and the video channel data;a single channel fingerprint component configured to:ngenerate a set of audio fingerprints based on the audio spectrogram,generate a set of wavelet min-hashes based on the set of mean frames, andgenerate a set of video fingerprints based on the set of wavelet min-hashes,wherein video fingerprints in the set of video fingerprints are min-hashes;anda multi-channel fingerprint component configured to generate a set of multi-channel fingerprints based on the set of audio fingerprints and the set of video fingerprints.

A method comprising:nreceiving, by a system including a processor, audio channel data and video channel data associated with a video;generating, by the system, an audio spectrogram based on the audio channel data; andgenerating, by the system, a set of interest points based on the audio spectrogram; andgenerating, by the system, a set of pairs wherein each pair in the set of pairs contains an anchor interest point and a paired interest point;generating, by the system, a third point for each pair in the set of pairs based on a search path, wherein the third point is a time-frequency point of a maxima along the search path;generating, by the system, a set of triples, wherein respective triples in the set of triples comprise the anchor interest point of a pair, the paired interest point of the pair, and the third point for the pair;determining, by the system, a binary bit associated with each triple in the set of triples based on whether the third point lies on a first half of the search path or a second half of the search path;generating, by the system, a set of descriptors based on the set of triples, wherein respective descriptors comprise a quantized frequency of an anchor interest point of a triple, a first quantized frequency ratio of a frequency of a paired interest point of the triple and a frequency of the anchor interest point of the triple, a second quantized frequency ratio of a frequency of a third point of the triple and the frequency of the anchor interest point of the triple, a time span between the anchor interest point of the triple and the paired interest point of the triple, and a binary bit associated with the triple;generating, by the system, a set of audio fingerprints based on the set of descriptors;generating, by the system, a set of video fingerprints based on the video channel data; andgenerating, by the system, a set of multi-channel fingerprints based on the set of audio fingerprints and the set of video fingerprints.

A system comprising:na processor; anda memory communicatively coupled to the processor, the memory having stored therein computer-executable instructions, comprising:nan input component configured to receive audio channel data and video channel data associated with a video;a spectrogram component configured to generate an audio spectrogram based on the audio channel data;an interest point detection component configured to generate a set of interest points based on the audio spectrogram;a pairing component configured to generate a set of pairs wherein each pair in the set of pairs contains an anchor interest point and a paired interest point;a third point component configured to generate a third point for each pair in the set of pairs based on a search path, wherein the third point is a time-frequency point of a maxima along the search path;a triples component configured to generate a set of triples, wherein respective triples in the set of triples comprise the anchor interest point of a pair, the paired interest point of the pair, and the third point for the pair;a search path component configured to determine a binary bit associated with the triple based on whether the third point lies on a first half of the search path or a second half of the search path;a descriptor component configured to generate a set of descriptors based on the set of triples, wherein respective descriptors comprise a quantized frequency of an anchor interest point of a triple, a first quantized frequency ratio of a frequency of a paired interest point of the triple and a frequency of the anchor interest point of the triple, a second quantized frequency ratio of a frequency of a third point of the triple and the frequency of the anchor interest point of the triple, a time span between the anchor interest point of the triple and the paired interest point of the triple, and a binary bit associated with the triple;a single channel fingerprint component configured to generate a set of audio fingerprints based on the set of descriptors, and generate a set of video fingerprints based on the video channel data; anda multi-channel fingerprint component configured to generate a set of multi-channel fingerprints based on the set of audio fingerprints and the set of video fingerprints.