Loading patent…
AssigneeDISH Network Technologies India Private Limited
InventorsVikram Balaraja Shetty, Manasa Jami, Visali Manoharan, Dileep Puramana
Estimating the term…
The present disclosure is directed to techniques for generating recommendations based on visual content. A receiver device can receive a video stream comprising a set of frames, select at least one frame from the set of frames, and apply a computer vision model to the selected frame to extract object metadata for an object of interest. The object metadata can relate to at least one of a scene, person, character, channel identifier, or product identifier. Using the object metadata, the receiver device can execute a trained neural network to generate a set of recommended items. The trained neural network can be trained on program metadata, channel metadata, and/or scheduling metadata. The techniques can include generating and displaying an interactive program guide (IPG) that includes at least one item from the set of recommended items.
A method for dynamically processing visual content to generate content recommendations, the method comprising:
One or more computer-readable media having instructions stored thereon that, when executed by at least one processor, cause a computing system to perform operations for dynamically processing visual content to generate content recommendations, the operations comprising:
A computing system configured to perform operations for dynamically processing visual content to generate content recommendations, the operations comprising:
Dependencies are inferred from the claim text.