Systems and methods for automated video classification

1. A computer-implemented method comprising:

receiving, by a computing system, a set of video frames associated with at least a portion of a video;

identifying, by the computing system, for each video frame of the set of video frames, a plurality of interest points based on an interest point detector;

determining, by the computing system, for each video frame of the set of video frames, that the video frame depicts the same static image as a next video frame in the set of video frames based on the plurality of interest points identified in each video frame;

classifying, by the computing system, the video as a static image video based on a determination that a threshold number of video frames from the set of video frames depict the same static image; and

downranking, by the computing system, the video in a content ranking system based on classification of the video as the static image video.

2. The computer-implemented method of claim 1 , wherein the determining, for each video frame of the set of video frames, whether the video frame depicts the same static image as a next video frame comprises determining whether the next video frame comprises a threshold number of interest points in common with the video frame.

3. The computer-implemented method of claim 2 , wherein the determining, for each video frame of the set of video frames, whether the video frame depicts the same static image as a next video frame further comprises determining whether each interest point in common between the video frame and the next video frame has moved less than a threshold distance from the video frame to the next video frame.

4. The computer-implemented method of claim 3 , wherein the threshold distance is a number of pixels.

5. The computer-implemented method of claim 1 , wherein the video comprises a plurality of video frames, and the set of video frames comprises a sampled subset of the plurality of video frames.

6. The computer-implemented method of claim 1 , wherein the video comprises a static image and one or more semi-transparent moving objects overlaid on the static image.

7. The computer-implemented method of claim 1 , wherein the determining, for each video frame of the set of video frames, whether the video frame depicts the same static image as a next video frame comprises:

determining that a first set of consecutive video frames from the set of video frames depicts a first static image, and

determining that a second set of consecutive video frames from the set of video frames depicts a second static image.

8. The computer-implemented method of claim 7 , further comprising:

identifying the first set of consecutive video frames as a first suspicious segment based on a determination that the first set of consecutive video frames comprises at least a threshold number of consecutive video frames, and

identifying the second set of consecutive video frames as a second suspicious segment based on a determination that the second set of consecutive video frames comprises at least the threshold number of consecutive video frames.

9. The computer-implemented method of claim 8 , further comprising classifying the video as a slideshow video based on a determination that the set of video frames comprises a threshold number of suspicious segments.

10. The computer-implemented method of claim 1 , further comprising:

determining, by the computing system, a subject associated with an interest point of the plurality of interest points; and

assigning, by the computing system, an interest point signature to the interest point.

11. A system comprising:

at least one processor; and

a memory storing instructions that, when executed by the at least one processor, cause the system to perform a method comprising:

receiving a set of video frames associated with at least a portion of a video;

identifying, for each video frame of the set of video frames, a plurality of interest points based on an interest point detector;

determining, for each video frame of the set of video frames, that the video frame depicts the same static image as a next video frame in the set of video frames based on the plurality of interest points identified in each video frame;

classifying the video as a static image video based on a determination that a threshold number of video frames from the set of video frames depict the same static image; and

downranking the video in a content ranking system based on classification of the video as the static image video.

12. The system of claim 11 , wherein the determining, for each video frame of the set of video frames, whether the video frame depicts the same static image as a next video frame comprises determining whether the next video frame comprises a threshold number of interest points in common with the video frame.

13. The system of claim 12 , wherein the determining, for each video frame of the set of video frames, whether the video frame depicts the same static image as a next video frame further comprises determining whether each interest point in common between the video frame and the next video frame has moved less than a threshold distance from the video frame to the next video frame.

14. The system of claim 13 , wherein the threshold distance is a number of pixels.

15. The system of claim 11 , wherein the instructions cause the system to perform the method further comprising:

determining a subject associated with an interest point of the plurality of interest points; and

assigning an interest point signature to the interest point.

16. A non-transitory computer-readable storage medium including instructions that, when executed by at least one processor of a computing system, cause the computing system to perform a method comprising:

receiving a set of video frames associated with at least a portion of a video;

identifying, for each video frame of the set of video frames, a plurality of interest points based on an interest point detector;

determining, for each video frame of the set of video frames, that the video frame depicts the same static image as a next video frame in the set of video frames based on the plurality of interest points identified in each video frame;

classifying the video as a static image video based on a determination that a threshold number of video frames from the set of video frames depict the same static image; and

downranking the video in a content ranking system based on classification of the video as the static image video.

17. The non-transitory computer-readable storage medium of claim 16 , wherein the determining, for each video frame of the set of video frames, whether the video frame depicts the same static image as a next video frame comprises determining whether the next video frame comprises a threshold number of interest points in common with the video frame.

18. The non-transitory computer-readable storage medium of claim 17 , wherein the determining, for each video frame of the set of video frames, whether the video frame depicts the same static image as a next video frame further comprises determining whether each interest point in common between the video frame and the next video frame has moved less than a threshold distance from the video frame to the next video frame.

19. The non-transitory computer-readable storage medium of claim 18 , wherein the threshold distance is a number of pixels.

20. The non-transitory computer-readable storage medium of claim 16 , wherein the instructions cause the computing system to perform the method further comprising:

determining a subject associated with an interest point of the plurality of interest points; and

assigning an interest point signature to the interest point.