Image recognition of content
Techniques are described to employ image recognition techniques to content. In an implementation, one or more images are identified in content using a signature derived from the one or more images. Metadata associated with the content is then supplemented based on the identified one or more images.
1. A method implemented by a computing device, the method comprising:
identifying that an image includes at least part of a weapon, the image included in one of a plurality of frames that form a segment of content; and
blocking output of each of the plurality of frames that form the segment of content based on the identification.
2. A method as described in claim 1 , wherein the image further includes material determined to be potentially harmful to a child.
3. A method as described in claim 1 , wherein the image further includes a corporate logo.
4. A client device comprising one or more modules at least partially implemented in hardware and configured to navigate between a plurality of segments of content based on identification by the client device of an image in each of the segments, the segments formed to include a plurality of frames of content of a television program, the navigating configured to support jumping from a segment of the content that includes the identified one or more images to another segment of the content that includes the identified one or more images without outputting another segment of the content that does not include the identified one or more images.
5. A client device as described in claim 4 , wherein the navigation is performed between segments that include at least one image, respectively, of a common object.
6. A client device as described in claim 4 , wherein the other segment of the content that does not include the identified one or more images and is disposed sequentially between the segments of the content that include the identified one or more images.
7. A method implemented by a computing device, the method comprising:
identifying that an image includes at least part of a corporate logo, the image included in a segment that includes a plurality of frames of content in a broadcast; and
blocking output of the segment of content that includes the plurality of frames of the content in the broadcast based on the identification.
8. A method as described in claim 7 , wherein the image further includes material determined to be potentially harmful to a child.
9. A method as described in claim 7 , wherein the image further includes at least a part of a weapon.