Placing sponsored-content based on images in video content
View Patent ↗Sponsored-content may be placed based on images in video content. A first image in a frame of a video content item is identified. The first image is matched with a second stored image. A sponsored-content item to be presented is selected based on an association between the second stored image and the sponsored-content item.
1. A method comprising:
selecting, from a buffer, a first frame of a video content item;
selecting a region of interest of the video content item visible in the first frame;
identifying a first image in the region of interest;
using object recognition techniques to match, by a computer, the first image received in a request for an advertisement with a second stored image, the second stored image being directly associated with one or more sponsored-content items;
selecting, by the computer, a sponsored-content item from the one or more sponsored-content items to be presented based on an association between the second stored image and the sponsored-content item;
selecting, based on a dynamic sampling rate that depends on a number of frames in the buffer, a second frame of the video content item;
comparing the second frame of the video content item with the first frame of the video content item to determine whether the first and second frames are statistically similar; and
upon determining that the first and second frames are statistically similar, associating the first image from the first frame with the second frame and selecting an advertisement based on the first image from the first frame.
2. The method of claim 1 further comprising storing the sponsored-content item in association with the video content item for later retrieval and display while the video content item is being played.
3. The method of claim 1 further comprising:
playing the video content item; and
displaying the sponsored-content item within or adjacent to the video content item.
4. The method of claim 1 wherein matching comprises identifying that the first image that is identical to the second image.
5. The method of claim 1 wherein matching comprises identifying that the first image that is substantially similar to the second image.
6. The method of claim 1 wherein matching the first image comprises matching a sub-image of the first image with a second sub-image of the second image.
7. The method of claim 6 wherein matching the sub-image of the first image comprises:
using object recognition techniques to generate a likelihood that the second sub-image includes the first sub-image of the first image; and
wherein selecting a sponsored-content item comprises selecting a sponsored-content item to be presented based on a comparison between the likelihood and a threshold.
8. The method of claim 1 wherein selecting a sponsored-content item comprises selecting a sponsored-content item to be presented based on both a bid associated with the sponsored-content item and the association between the second stored image and the sponsored-content item.
9. The method of claim 1 wherein selecting a sponsored-content item comprises selecting a sponsored-content item to be presented based on both a maximum bid associated with the sponsored-content item and the association between the second stored image and the sponsored-content item.
10. The method of claim 1 , wherein the association between the second stored image and the sponsored-content item comprises a pointer.
11. The method of claim 1 , wherein the association between the second stored image and the sponsored-content item comprises a link.
12. The method of claim 1 , wherein the association between the second stored image and the sponsored-content item comprises an association stored in a data store between the sponsored-content item and the image.
13. The method of claim 1 , wherein matching the first image with the second stored image comprises matching the first image with the second stored image while the video is playing.
14. The method of claim 1 , further comprising:
playing the video; and
displaying the advertisement while the video is playing.
15. A computer system comprising instructions, that when executed by the computer system, cause the computer system to perform operations comprising:
selecting, from a buffer, a first frame of a video content item;
selecting a region of interest of the video content item visible in the first frame;
identifying a first image in the region of interest;
using object recognition techniques to match the first image received in a request for an advertisement with a second stored image, the second stored image being directly associated with one or more sponsored-content items;
selecting a sponsored-content item from the one or more sponsored-content items to be presented based on an association between the second stored image and the sponsored-content item;
selecting, based on a dynamic sampling rate that depends on a number of frames in the buffer, a second frame of the video content item;
comparing the second frame of the video content item with the first frame of the video content item to determine whether the first and second frames are statistically similar; and
upon determining that the first and second frames are statistically similar, associating the first image from the first frame with the second frame and selecting an advertisement based on the first image from the first frame.
16. The computer system of claim 15 further comprising instructions to cause the computer to store the sponsored-content item in association with the video content item for later retrieval and display while the video content item is being played.
17. The computer system of claim 15 further comprising instructions to cause the computer to:
play the video content item; and
display the sponsored-content item within or adjacent to the video content item.
18. The computer system of claim 15 wherein matching comprises identifying that the first image that is identical to the second image.
19. The computer system of claim 15 wherein matching comprises identifying that the first image that is substantially similar to the second image.
20. The computer system of claim 15 wherein matching the first image comprises matching a sub-image of the first image with a second sub-image of the second image.
21. The computer system of claim 20 wherein matching the sub-image of the first image comprises:
using object recognition techniques to generate a likelihood that the second sub-image includes the first sub-image of the first image; and
wherein selecting a sponsored-content item comprises selecting a sponsored-content item to be presented based on a comparison between the likelihood and a threshold.
22. The computer system of claim 15 wherein selecting a sponsored-content item comprises selecting a sponsored-content item to be presented based on both a bid associated with the sponsored-content item and the association between the second stored image stored and the sponsored-content item.