Identification of captured videos

摘要

An identification system receives a request from a user device to identify a video captured in a video stream. The identification system identifies a display area captured in the video stream in which the video is displayed. The identification system identifies content include in the display area throughout the stream. Based on the identified content, the identification system creates a fingerprint for the video. The identification system identifies from a plurality of stored fingerprints at least one fingerprint similar to the created fingerprint. The identification system identifies video information associated with the similar fingerprint. The video information is transmitted to the user device and displayed as being video information about the video captured in the stream.

主权项

1. A computer-implemented method comprising:
receiving, by a computer system from a first device, a request to identify a video; receiving, by the computer system from the first device, a video stream captured by the first device, the video stream capturing a second device displaying the video; identifying, by the computer system from the video stream, a plurality of candidate edges, the plurality of candidate edges including more than four edges; responsive to identifying the plurality of candidate edges, selecting, by the computer system from the plurality of candidate edges, four edges enclosing a display area in the video stream corresponding to the second device, each of the four edges selected based on brightness levels of a plurality of pixels corresponding to the edge; creating, by the computer system, a fingerprint for the video based on content included in the display area; identifying, by the computer system, video information based on the created fingerprint; and transmitting, by the computer system to the first device, instructions to display the video information.