Content recognition for targeting video advertisements

1. A computer-implemented method, the method comprising;

identifying a plurality of advertisement targeting criteria based on content of a video stream or file, in which each advertisement targeting criterion describes content of a respective portion of the video at a respective time position in the video;

associating a first advertisement targeting criterion with a first time position in the video, wherein the first advertisement targeting criterion is selected for association with the first time position in the video based on a determination that the first time position occurs within a time interval of a plurality of non-overlapping time intervals of the video wherein each time interval is associated with a respective plurality of the identified advertisement targeting criteria greater than a predetermined threshold that are the same as, or related to, the first advertisement targeting criterion; and

providing one or more first advertisements based on the first advertisement targeting criterion for presentation at or after the first time position in the video.

2. The method of claim 1 wherein providing one or more first advertisements comprises:

providing the video and the first advertisement targeting criterion to a user device;

receiving from the user device a request for advertisements, the request including the first advertisement targeting criterion;

identifying one or more advertisements based on the first advertisement targeting criterion included in the request; and

providing the identified advertisements to the user device.

3. The method of claim 1 wherein identifying the plurality of advertisement targeting criteria comprises analyzing audio content of the video.

4. The method of claim 3 wherein analyzing the audio content of the video comprises applying speech recognition to the audio content of the video to identify speech in the audio content of the video; and

determining one or more advertisement targeting criteria from the identified speech.

5. The method of claim 1 wherein identifying the plurality of advertisement targeting criteria comprises analyzing visual content of the video.

6. The method of claim 5 wherein analyzing the visual content of the video comprises:

applying automatic object recognition to the visual content of the video to identify one or more objects in the visual content of the video; and

determining one or more advertisement targeting criteria from the identified objects.

7. The method of claim 1 wherein identifying the plurality of advertisement targeting criteria comprises:

extracting textual data from the video, wherein the textual data represents speech in the video; and

determining one or more advertisement targeting criteria from the textual data.

8. A system comprising:

a non-transitory computer-readable medium having instructions stored thereon; and

one or more processors programmed to execute the instructions to perform operations comprising:

identifying a plurality of advertisement targeting criteria based on content of a video stream or file, in which each advertisement targeting criterion describes content of a respective portion of the video at a respective time position in the video;

associate a first advertisement targeting criterion with a first time position in the video, wherein the first advertisement targeting criterion is selected for association with the first time position in the video based on a determination that the first time position occurs within a time interval of a plurality of non-overlapping time intervals of the video wherein each time interval is associated with a respective plurality of the identified advertisement targeting criteria greater than a predetermined threshold that are the same as, or related to, the first advertisement targeting criterion; and

provide one or more first advertisements based on the criterion for presentation at or after the first time position in the video.

9. A storage device having instructions stored thereon that, when executed by data processing apparatus, are operable to cause the data processing apparatus to perform operations comprising;

identifying a plurality of advertisement targeting criteria based on content of a video stream or file, in which each advertisement targeting criterion describes content of a respective portion of the video at a respective time position in the video described by the criterion;

associating a first advertisement targeting criterion with a first time position in the video, wherein the first advertisement targeting criterion is selected for association with the first time position in the video based on a determination that the first time position occurs within a time interval of a plurality of non-overlapping time intervals of the video wherein each time interval is associated with a respective plurality of the identified advertisement targeting criteria greater than a predetermined threshold that are the same as, or related to, the first advertisement targeting criterion; and

providing one or more first advertisements based on the first advertisement targeting criterion for presentation at or after the first time position in the video.

10. The system of claim 8 wherein providing one or more first advertisements comprises:

providing the video and the first advertisement targeting criterion to a user device;

receiving from the user device a request for advertisements, the request including the first advertisement targeting criterion;

identifying one or more advertisements based on the first advertisement targeting criterion included in the request; and

providing the identified advertisements to the user device.

11. The system of claim 8 wherein identifying the plurality of advertisement targeting criteria comprises analyzing audio content of the video.

12. The system of claim 11 wherein analyzing the audio content of the video comprises

applying speech recognition to the audio content of the video to identify speech in the audio content of the video; and

determining one or more advertisement targeting criteria from the identified speech.

13. The system of claim 8 wherein identifying the plurality of advertisement targeting criteria comprises analyzing visual content of the video.

14. The system of claim 13 wherein analyzing the visual content of the video comprises:

applying automatic object recognition to the visual content of the video to identify one or more objects in the visual content of the video; and

determining one or more advertisement targeting criteria from the identified objects.

15. The system of claim 8 wherein identifying the plurality of advertisement targeting criteria comprises:

extracting textual data from the video, wherein the textual data represents speech in the video; and

determining one or more advertisement targeting criteria from the textual data.

16. The storage device of claim 9 wherein providing one or more first advertisements comprises:

providing the video and the first advertisement targeting criterion to a user device;

receiving from the user device a request for advertisements, the request including the first advertisement targeting criterion;

identifying one or more advertisements based on the first advertisement targeting criterion included in the request; and

providing the identified advertisements to the user device.

17. The storage device of claim 9 wherein identifying the plurality of advertisement targeting criteria comprises analyzing audio content of the video.

18. The storage device of claim 17 wherein analyzing the audio content of the video comprises:

applying speech recognition to the audio content of the video to identify speech in the audio content of the video; and

determining one or more advertisement targeting criteria from the identified speech.

19. The storage device of claim 9 wherein identifying the plurality of advertisement targeting criteria comprises analyzing visual content of the video.

20. The storage device of claim 19 wherein analyzing the visual content of the video comprises:

applying automatic object recognition to the visual content of the video to identify one or more objects in the visual content of the video; and

determining one or more advertisement targeting criteria from the identified objects.

21. The storage device of claim 9 wherein identifying the plurality of advertisement targeting criteria comprises:

extracting textual data from the video, wherein the textual data represents speech in the video; and

determining one or more advertisement targeting criteria from the textual data.