Adaptive video zoom

1. On a computing device, a method of providing a video content item for viewing by one or more viewers, the method comprising:

identifying one or more objects within the video content item;

for each object identified within the video content item, assigning a searchable descriptor for the object;

for each object identified within the video content item, creating a time stamp indicating one or more time periods within the video content item during which the object occurs;

for each object identified within the video content item, creating a bounding window defining a region of the video content item within which the object occurs;

for each object identified within the video content item, packaging the searchable descriptor, the time stamp, and the bounding window into metadata associated with the video content item, the metadata facilitating zoomed viewing of a selected object of the one or more objects within the video content item; and

delivering the video content item and the metadata associated with the video content item to a client computing device.

2. The method of claim 1 , where the metadata is encoded as part of the video content item.

3. The method of claim 1 , where the metadata is a companion file distinct from the video content item.

4. The method of claim 1 , where the bounding window for each object indicates a zooming level for displaying the object.

5. The method of claim 1 , where a shape of the bounding window for each object is dependent on a shape of the object.

6. The method of claim 1 , where the searchable descriptor includes a searchable name.

7. The method of claim 1 , where identifying the one or more objects comprises applying a facial recognition algorithm.

8. The method of claim 1 , where identifying the one or more objects comprises applying an object recognition algorithm.

9. The method of claim 1 , wherein the computing device and the client computing device are of a same computing device.

10. On a client computing device, a method of viewing a video content item, the method comprising:

receiving a video content item and metadata associated with the video content item, the metadata comprising, for each of one or more objects within the video content item, a searchable descriptor for the object, a time stamp indicating one or more time periods within the video content item during which the object occurs, and a bounding window defining a region of the video content item within which the object occurs;

accessing a zoom preference of a viewer, the zoom preference indicating one or more objects to zoom during video playback;

selecting a selected object from the one or more objects by searching the metadata for an object indicated by the zoom preference; and

displaying a zoomed view of the selected object on a display of the client computing device.

11. The method of claim 10 , where the zoom preference includes zoom preferences identified by the viewer.

12. The method of claim 10 , where the zoom preference includes learned zoom preferences derived from past viewing habits.

13. The method of claim 10 , where the zoom preference is stored locally at the client computing device.

14. The method of claim 10 , where the zoom preference is stored at a network-accessible computing device.

15. The method of claim 10 , further comprising displaying a mosaic view visually presenting one or more candidate objects to zoom during video playback, and where selecting the selected object comprises receiving selection of a candidate object from the mosaic view.

16. The method of claim 15 , where the mosaic view is displayed on a display of the client computing device.

17. The method of claim 15 , where the mosaic view is displayed on a display of a companion computing device.

18. The method of claim 10 , where searching the metadata comprises searching the searchable descriptor.

19. The method of claim 10 , where the zoomed view visually presents the selected object of the video content item based on a bounding window associated with the selected object.

20. A method, comprising:

receive receiving a video content item and metadata associated with the video content item, the metadata comprising, for each of one or more objects within the video content item, a searchable descriptor for the object, a time stamp indicating one or more time periods within the video content item during which the object occurs, and a bounding window defining a region of the video content item within which the object occurs;

access accessing a zoom preference of a viewer, the zoom preference indicating one or more objects to zoom during video playback;

selecting a selected object from the one or more objects by searching the metadata for an object indicated by the zoom preference;

selecting on a computing device a first zoomed view of the selected object based on the zoom preference; and

displaying on a companion computing device a second zoomed view of the video content item based on the zoom preference.