Adaptive content classification of a video content item

1. A method, performed by one or more processors, for performing content classification of a video content item, the method comprising:

extracting a plurality of frames from the video content item;

combining the plurality of frames into single images, each single image comprising a collage of frames; and

analyzing the single images to identify a type of content in the single images,

wherein the extracting of a plurality of frames from the video content item comprises extracting frames at a sampling rate that dictates a frequency at which frames are included in the single images, the method further comprising:

responsive to identifying content within at least one of the single images indicative of the type of content, increasing the sampling rate of the frames; and

responsive to not identifying content within at least one of the single images indicative of the type of content, decreasing the sampling rate of the frames.

2. The method of claim 1 , wherein the sampling rate depends on the particular type of content.

3. The method of claim 1 , wherein the single images comprise two by two collages of frames or three by three collages of frames.

4. The method of claim 1 , wherein the analyzing of the single images comprises: applying at least one classifier to the single images, wherein the at least one classifier is configured to automatically identify a particular type of content.

5. The method of claim 4 , wherein a sensitivity of each of the at least one classifier depends on the particular type of content.

6. A method, performed by one or more processors, for performing content classification of a video content item, the method comprising:

extracting a plurality of frames from the video content item;

combining the plurality of frames into single images, each single image comprising a collage of frames; and

analyzing the single images to identify a type of content in the single images,

wherein the analyzing of the single images comprises:

sequentially applying a plurality of classifiers to the single image, wherein each classifier of the plurality of classifiers is configured to automatically identify a particular type of content.

7. The method of claim 1 , wherein the analyzing of the single images comprises: concurrently applying a plurality of classifiers to the single images, wherein each classifier of the plurality of classifiers is configured to automatically identify a particular type of content.

8. A method, performed by one or more processors, for performing content classification of a video content item, the method comprising:

extracting a plurality of frames from the video content item;

combining the plurality of frames into single images, each single image comprising a collage of frames;

analyzing the single images to identify a type of content in the single images;

determining that it is inconclusive that the video content item comprises the type of content; and

forwarding the video content item for human review,

wherein the determination that it is inclusive is based on a numeric value representing a likelihood of the video content including the type of content.

9. A non-transitory computer readable storage medium having computer readable program code stored thereon for causing a computer system to perform a method for performing content classification of a video content item, the method comprising:

extracting a plurality of frames from the video content item;

combining the plurality of frames into single images, each single image comprising a collage of frames; and

analyzing the single images to identify a type of content in the single images, wherein the extracting of a plurality of frames from the video content item comprises extracting frames at a sampling rate that dictates a frequency at which frames are included in the single images, the method further comprising:

responsive to identifying content within at least one of the single images indicative of the type of content, increasing the sampling rate of the frames; and

responsive to not identifying content within at least one of the single images indicative of the type of content, decreasing the sampling rate of the frames.

10. A non-transitory computer readable storage medium having computer readable program code stored thereon for causing a computer system to perform a method for performing content classification of a video content item, the method comprising:

extracting a plurality of frames from the video content item;

combining the plurality of frames into single images, each single image comprising a collage of frames; and

analyzing the single images to identify a type of content in the single images,

wherein the analyzing of the single images comprises:

sequentially applying a plurality of classifiers to the single image, wherein each classifier of the plurality of classifiers is configured to automatically identify a particular type of content.

11. The non-transitory computer readable storage medium of claim 9 wherein the sampling rate depends on the particular type of content.

12. The non-transitory computer readable storage medium of claim 9 , the method further comprising:

applying at least one classifier to the single images, wherein the at least one classifier is configured to automatically identify a particular type of content.

13. The non-transitory computer readable storage medium of claim 12 , wherein a sensitivity of the at least one classifier depends on the particular type of content.

14. The non-transitory computer readable storage medium of claim 9 , the method further comprising:

determining that it is inconclusive that the video content item comprises the type of content; and

forwarding the video content item for human review.

15. A computer system comprising:

a data storage unit; and

at least one processor coupled with the data storage unit, the at least one processor being configured to perform a method for performing content classification of a video content item, the method comprising:

extracting a plurality of frames from the video content item;

combining the plurality of frames into single images, each single image comprising a collage of frames; and

analyzing the single images to identify a type of content in the single images, wherein the extracting of a plurality of frames from the video content item comprises extracting frames at a sampling rate that dictates a frequency at which frames are included in the single images, the method further comprising:

responsive to identifying content within at least one of the single images indicative of the type of content, increasing the sampling rate of the frames; and

responsive to not identifying content within at least one of the single images indicative of the type of content, decreasing the sampling rate of the frames.

16. A computer system comprising:

a data storage unit; and

at least one processor coupled with the data storage unit, the at least one processor being configured to perform a method for performing content classification of a video content item, the method comprising:

extracting a plurality of frames from the video content item;

combining the plurality of frames into single images, each single image comprising a collage of frames;

analyzing the single images to identify a type of content in the single images;

determining that it is inconclusive that the video content item comprises the type of content; and

forwarding the video content item for human review, wherein the determination that it is inclusive is based on a numeric value representing a likelihood of the video content including the type of content.

17. The computer system of claim 15 , wherein the sampling rate depends on the particular type of content.

18. The computer system of claim 15 , the method further comprising:

applying a classifier to the single images,

wherein the classifier is configured to automatically identify a particular type of content and a sensitivity of the classifier depends on the particular type of content.