Video fingerprinting based on fourier transform of histogram
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.
1. A content device comprising:
a non-transitory memory storing instructions; and
a processing device in communication with the non-transitory memory and configured to execute the instructions, the executed instructions causing the processing device to perform operations comprising:
receiving streaming video content from a content provider;
capturing a frame of the streaming video content; and
fingerprinting the captured frame of the streaming video content by:
generating a one-dimensional histogram function of pixel values based on pixel data for an entirety of the captured frame, the one-dimensional histogram function comprising an X axis representing specific pixel values and a Y axis representing a number of pixels at each respective specific pixel value;
transforming the histogram function with a Fast Fourier Transform (FFT) to generate a plurality of complex values for the captured frame;
for each of the plurality of complex values, generating a corresponding bit by assigning a binary one (“1”) when a real part of the respective complex value is greater than zero (“0”) and assigning a binary zero (“0”) when the real part of the respective complex value is less than or equal to zero; and
concatenating a number of the bits to generate a fingerprint for the captured frame.
2. The content device of claim 1 , wherein each specific pixel value of the histogram function comprises a corresponding Y value of a YUV color space.
3. The content device of claim 1 , wherein the operations further comprise, when fingerprinting the captured frame, indexing the plurality of complex values from zero (“0”) to 255.
4. The content device of claim 1 , wherein the generated fingerprint is 128 bits long or 256 bits long.
5. The content device of claim 1 , wherein the operations further comprise communicating the generated fingerprint to an automatic content recognition (ACR) server configured to store the generated fingerprint in relation to corresponding broadcast schedule metadata.
6. The content device of claim 1 , wherein the streaming video comprises encoded video content.
7. The content device of claim 1 , wherein the streaming video comprises raw video within a YUV 4:2:2 color space.
8. The content device of claim 1 , wherein the operations further comprise converting the streaming video by scaling a resolution of the streaming video to a threshold resolution level.
9. The content device of claim 8 , wherein the converting operation further comprises converting the streaming video to YUV 4:2:0 color space.
10. A method comprising:
receiving, at a processing device of a content device, streaming video content from a content provider;
capturing, by the processing device, a frame of the streaming video content; and
fingerprinting, by the processing device, the captured frame of the streaming video content by:
generating a one-dimensional histogram function of pixel values based on pixel data for an entirety of the captured frame, the one-dimensional histogram function comprises an X axis representing specific pixel values and a Y axis representing a number of pixels at each respective specific pixel value;
transforming the histogram function with a Fast Fourier Transform (FFT) to generate a plurality of complex values for the captured frame;
for each of the plurality of complex values, generating a corresponding bit by assigning a binary one (“1”) when a real part of the respective complex value is greater than zero (“0”) and assigning a binary zero (“0”) when the real part of the respective complex value is less than or equal to zero; and
concatenating a number of the bits to generate a fingerprint for the captured frame.
11. The method of claim 10 , wherein each specific pixel value of the histogram function comprises a corresponding Y value of a YUV color space.
12. The method of claim 10 , wherein when fingerprinting the captured frame, indexing the plurality of complex values from zero (“0”) to 255.
13. The method of claim 10 , wherein the generated fingerprint is 128 bits long or 256 bits long.
14. The method of claim 10 , further comprising communicating, by the processing device, the generated fingerprint to an automatic content recognition (ACR) server configured to store the generated fingerprint in relation to corresponding broadcast schedule metadata.
15. The method of claim 10 , wherein the streaming video comprises encoded video content.
16. The method of claim 10 , wherein the streaming video comprises raw video within a YUV 4:2:2 color space.
17. The method of claim 10 , wherein fingerprinting, by the processing device, further comprises converting the streaming video by scaling a resolution of the streaming video to a threshold resolution level.
18. The method of claim 17 , wherein converting further comprises converting the streaming video to YUV 4:2:0 color space.