Systems and methods for performing a search based on a media content snapshot image

摘要

Systems and methods are provided for performing a search based on a snapshot image captured from media content presented to a user. The snapshot image contains features of the media content that the user wishes to target for the search. A search system recognizes features of the snapshot image and creates a search query based on the snapshot image. The search query is used to identify features of the snapshot image, and search results related to the identified features are presented to the user. Supplemental data or user input received with the snapshot image may be used in analyzing and identifying features of the snapshot image.

主权项

1. A method for performing an image-based search, the method comprising:
receiving a snapshot image captured from video of media content being displayed by user equipment; receiving supplemental data associated with the media content; retrieving a plurality of images from a collection of searchable data by comparing the supplemental data to data associated with each image of the plurality of images; processing visual attributes of each image of the plurality of retrieved images to determine whether at least one image of the plurality of retrieved images includes at least one recognizable feature of the snapshot image; providing at least one first search result based on determining at least one image of the plurality of retrieved images has the at least one recognizable feature; receiving a user request to refine the image-based search; processing at least one new image based at least in part on the user request; and providing at least one second search result based on the at least one new image.