| 摘要 |
Images may be automatically annotated with place identifiers based on textual metadata associated with the images. The quality of these place identifiers may be scored based on a number of different factors. For example, these annotations may also include confidence and topicality values indicative of the relationships between the place identifier and the text as well as topics of the text of the textual metadata. Images that also have additional location information that includes GPS coordinates or a location specified by a user, may be used to evaluate the quality of a given place identifier, combination of confidence and topicality values, and the overall annotation system. These valuations may, in turn, be used to score the quality of such automatic annotations of other images that do not have such additional location information. |
| 主权项 |
1. A computer implemented method for selecting time-distributed panoramas for display, the method comprising:
accessing, by one or more computing devices, a plurality of panoramic images, each given panoramic image of the plurality of panoramic images being associated with multi-dimensional location information defined in at least two dimensions, uni-dimensional (1D) location information defined in only one dimension, three dimensional (3D) geometry data corresponding to a surface of an object depicted in the given panoramic image, and time information identifying when the given panoramic image was captured; for a first one of the plurality of panoramic images, the first panoramic image being associated with first multi-dimensional location information, first 1D location information, and first time information, selecting, by the one or more computing devices, a second panoramic image of the plurality of panoramic images, the second panoramic image panoramic image being associated with second multi-dimensional location information, second 1D location information, and second time information, the second panoramic image being selected such that the second 1D location information is closest to the first 1D location information and the first time information and second time information indicate that the first panoramic image and the second panoramic image were captured on different days; providing, by the one or more computing devices, the first panoramic image for display with an indication that the second panoramic image is available; determining, by the one or more computing devices, a zoom level for the second panoramic image based on the first 3D geometry data, the first multi-dimensional location information, the second 3D geometry data, and the second multi-dimensional location information; and providing, by the one or more computing devices, the second panoramic image for display with the zoom level for the second panoramic image. |
