Sound and video object tracking
AssigneeDolby Laboratories Licensing Corporation
InventorsCong Zhou, Timo Kunkel, Cristina Michel Vasco
Life of this patent
Estimating the term…
Abstract
Image data relating to real-world objects or persons is collected from a scene while collecting audio data relating to the real-world objects or persons from the same scene. The audio data is used to derive sound objects corresponding to the real-world objects or persons. The image data is used to derive video objects corresponding to the real-world objects or persons. Based on the sound objects and the video objects, candidate salient objects are generated. A salient object is selected from among the candidate salient objects. Perceptual enhancement operations are performed on the selected salient object.
Claims19
A method, comprising:
- collecting image data relating to one or more real-world objects or persons from a scene while collecting audio data relating to the one or more real-world objects or persons from the same scene, the audio data being used to derive one or more sound objects corresponding to the one or more real-world objects or persons, the image data being used to derive one or more video objects corresponding to the one or more real-world objects or persons;
- generating, based on the one or more sound objects and the one or more video objects, one or more candidate salient objects;
- selecting a salient object from among the one or more candidate salient objects and performing one or more perceptual enhancement operations on the salient object;
- wherein the method is performed by one or more computing devices.
Dependencies are inferred from the claim text.