Scene frame matching for automatic content recognition

    公开(公告)号:US10922551B2

    公开(公告)日:2021-02-16

    申请号:US15726987

    申请日:2017-10-06

    发明人: Juikun Chen

    摘要: A method for scene frame matching for automatic content recognition. The method includes receiving fingerprints indicative of frames of a media stream. The method also includes obtaining scene centroids of corresponding scenes of the media stream. Each scene includes fingerprints ranging from an initial frame having a corresponding initial fingerprint and a last frame having a corresponding last fingerprint. Each scene centroid represents a mean of fingerprints corresponding to a respective scene. The mean of fingerprints is determined by an average of pixel values corresponding to the fingerprints of the respective scene. For each scene centroid, the method includes determining a distance between the respective scene centroid and a target fingerprint of a corresponding target frame of the media stream and determining a target scene of the target frame as being the respective scene of the scene centroid having a shortest distance to the target fingerprint of the target frame.

    Automatic content recognition with local matching

    公开(公告)号:US10750236B2

    公开(公告)日:2020-08-18

    申请号:US14813416

    申请日:2015-07-30

    发明人: Juikun Chen

    摘要: A media device and method execute computer instructions to: generate a first query fingerprint of media content; send the first query fingerprint to a server; and receive, from the server, an identification of a first channel being watched in response to the server finding a match to the first query fingerprint. The media device iteratively receives, from the server, a plurality of fingerprints generated from the media content in locations ahead of where the media content is being received; store the plurality of fingerprints in a local database of a computer-readable storage medium; and compare a second query fingerprint of the media content with the plurality of fingerprints to determine whether the media device remains set on the first channel, without continuously pinging the server. When the media device locally detects a non-match with the second query fingerprint, the process starts over with it finding a match at the server.

    Fingerprint layouts for content fingerprinting

    公开(公告)号:US10743062B2

    公开(公告)日:2020-08-11

    申请号:US16287752

    申请日:2019-02-27

    发明人: Juikun Chen

    摘要: A method is disclosed for identifying the media content. The method includes receiving fingerprint media content from a content provider to obtain a first set of fingerprints, the first set of fingerprints including a first subset of fingerprints in a first fingerprint layout and a second subset of fingerprints in a second fingerprint layout. The method can include receiving a second set of fingerprints for media content displayed on the electronic device, the second set of fingerprints comprising a third subset of fingerprints in the first fingerprint layout and a fourth subset of fingerprints in the second fingerprint layout. The method can include matching a first fingerprint in the first subset of fingerprints to a second fingerprint in the third subset of fingerprints or a third fingerprint in the second subset of fingerprints to a fourth fingerprint in the fourth subset of fingerprints in order to identify the media content.

    Video Fingerprinting Based on Fourier Transform of Histogram

    公开(公告)号:US20190327525A1

    公开(公告)日:2019-10-24

    申请号:US16502649

    申请日:2019-07-03

    发明人: Juikun Chen

    摘要: A content device and method is disclosed to include a processing device to process streaming video content. A fingerprinter receives captured frames of the streaming video content and, for each frame of a plurality of the captured frames, generates a one-dimensional histogram function of pixel values and transforms the histogram function with a Fast Fourier Transform (FFT), to generate a plurality of complex values for the frame. The fingerprinter further, for each of the plurality of complex values, assigns a binary one (“1”) when a real part of the complex value is greater than zero (“0”) and assigns a binary zero (“0”) when the real part is less than or equal to zero, to generate a plurality of bits. The fingerprinter further concatenates a specific number of the bits to generate a fingerprint for the frame.

    Scene Frame Matching for Automatic Content Recognition

    公开(公告)号:US20190303681A1

    公开(公告)日:2019-10-03

    申请号:US16447344

    申请日:2019-06-20

    发明人: Juikun Chen

    摘要: A method for scene frame matching for automatic content recognition. The method includes receiving fingerprints indicative of frames of a media stream. The method also includes obtaining scene centroids of corresponding scenes of the media stream. Each scene includes fingerprints ranging from an initial frame having a corresponding initial fingerprint and a last frame having a corresponding last fingerprint. Each scene centroid represents a mean of fingerprints corresponding to a respective scene. The mean of fingerprints is determined by an average of pixel values corresponding to the fingerprints of the respective scene. For each scene centroid, the method includes determining a distance between the respective scene centroid and a target fingerprint of a corresponding target frame of the media stream and determining a target scene of the target frame as being the respective scene of the scene centroid having a shortest distance to the target fingerprint of the target frame.

    Flagging advertisement frames for automatic content recognition

    公开(公告)号:US10887667B2

    公开(公告)日:2021-01-05

    申请号:US16409733

    申请日:2019-05-10

    发明人: Juikun Chen

    摘要: A method for flagging advertisement frames for automatic content recognition is provided. The method includes receiving broadcast fingerprints indicative of broadcast frames of a media stream comprising a series of broadcast scenes. The method also includes receiving advertisement fingerprints indicative of ad frames of ad scenes. The method further includes determining a scene change between a first broadcast scene and a second broadcast scene. The scene change is based on a Pearson correlation coefficient between an initial broadcast fingerprint of an initial broadcast frame of the second broadcast scene and a last broadcast fingerprint of a last broadcast frame of the first broadcast scene. The method also further includes determining whether the second broadcast scene is one of the ad scenes. When the second broadcast scene is one of the ad scenes, the method associates an identification of the second broadcast scene as the one of the ad scenes.

    AUTOMATIC CONTENT RECOGNITION WITH LOCAL MATCHING

    公开(公告)号:US20200344520A1

    公开(公告)日:2020-10-29

    申请号:US16924119

    申请日:2020-07-08

    发明人: Juikun Chen

    摘要: A media device and method execute computer instructions to: generate a first query fingerprint of media content; send the first query fingerprint to a server; and receive, from the server, an identification of a first channel being watched in response to the server finding a match to the first query fingerprint. The media device iteratively receives, from the server, a plurality of fingerprints generated from the media content in locations ahead of where the media content is being received; store the plurality of fingerprints in a local database of a computer-readable storage medium; and compare a second query fingerprint of the media content with the plurality of fingerprints to determine whether the media device remains set on the first channel, without continuously pinging the server. When the media device locally detects a non-match with the second query fingerprint, the process starts over with it finding a match at the server.

    Video fingerprinting based on fourier transform of histogram

    公开(公告)号:US10390083B2

    公开(公告)日:2019-08-20

    申请号:US15826044

    申请日:2017-11-29

    发明人: Juikun Chen

    摘要: A content device and method is disclosed to include a processing device to process streaming video content. A fingerprinter receives captured frames of the streaming video content and, for each frame of a plurality of the captured frames, generates a one-dimensional histogram function of pixel values and transforms the histogram function with a Fast Fourier Transform (FFT), to generate a plurality of complex values for the frame. The fingerprinter further, for each of the plurality of complex values, assigns a binary one (“1”) when a real part of the complex value is greater than zero (“0”) and assigns a binary zero (“0”) when the real part is less than or equal to zero, to generate a plurality of bits. The fingerprinter further concatenates a specific number of the bits to generate a fingerprint for the frame.

    Scene Frame Matching for Automatic Content Recognition

    公开(公告)号:US20210133460A1

    公开(公告)日:2021-05-06

    申请号:US17142619

    申请日:2021-01-06

    发明人: Juikun Chen

    摘要: A method for scene frame matching for automatic content recognition. The method includes receiving fingerprints indicative of frames of a media stream. The method also includes obtaining scene centroids of corresponding scenes of the media stream. Each scene includes fingerprints ranging from an initial frame having a corresponding initial fingerprint and a last frame having a corresponding last fingerprint. Each scene centroid represents a mean of fingerprints corresponding to a respective scene. The mean of fingerprints is determined by an average of pixel values corresponding to the fingerprints of the respective scene. For each scene centroid, the method includes determining a distance between the respective scene centroid and a target fingerprint of a corresponding target frame of the media stream and determining a target scene of the target frame as being the respective scene of the scene centroid having a shortest distance to the target fingerprint of the target frame.

    Flagging Advertisement Frames for Automatic Content Recognition

    公开(公告)号:US20210084386A1

    公开(公告)日:2021-03-18

    申请号:US17247070

    申请日:2020-11-27

    发明人: Juikun Chen

    摘要: A method for flagging advertisement frames for automatic content recognition is provided. The method includes receiving broadcast fingerprints indicative of broadcast frames of a media stream comprising a series of broadcast scenes. The method also includes receiving advertisement fingerprints indicative of ad frames of ad scenes. The method further includes determining a scene change between a first broadcast scene and a second broadcast scene. The scene change is based on a Pearson correlation coefficient between an initial broadcast fingerprint of an initial broadcast frame of the second broadcast scene and a last broadcast fingerprint of a last broadcast frame of the first broadcast scene. The method also further includes determining whether the second broadcast scene is one of the ad scenes. When the second broadcast scene is one of the ad scenes, the method associates an identification of the second broadcast scene as the one of the ad scenes.