Techniques for reconstructing downscaled video content
AssigneeNetflix, Inc.
InventorsChristos G. BAMPIS, Zhi LI
Life of this patent
Estimating the term…
Abstract
In various embodiments an endpoint application reconstructs downscaled videos. The endpoint application accesses metadata associated with a portion of a downscaled video that has a first resolution and was generated using a trained downscaling convolutional neural network (CNN). The endpoint application determines, based on the metadata, an upscaler that should be used when upscaling the portion of the downscaled video. The endpoint application executes the upscaler on the portion of the downscaled video to generate a portion of a reconstructed video that is accessible for playback and has a second resolution that is greater than the first resolution.
Claims20
A computer-implemented method for reconstructing downscaled videos, the method comprising:
- accessing metadata associated with a first portion of a first downscaled video that has a first resolution and was generated using a trained downscaling convolutional neural network (CNN);
- determining, based on the metadata, a first upscaler that should be used when upscaling the first portion of the first downscaled video;
- and
- executing the first upscaler on the first portion of the first downscaled video to generate a first portion of a first reconstructed video that is accessible for playback and has a second resolution that is greater than the first resolution.
One or more non-transitory computer readable media including instructions that, when executed by one or more processors, cause the one or more processors to reconstruct downscaled videos by performing the steps of:
- accessing metadata associated with a first portion of a first downscaled video that has a first resolution and was generated using a trained downscaling convolutional neural network (CNN);
- determining, based on the metadata, a first upscaler that should be used when upscaling the first portion of the first downscaled video;
- and
- executing the first upscaler on the first portion of the first downscaled video to generate a first portion of a first reconstructed video that is accessible for playback and has a second resolution that is greater than the first resolution.
A system comprising:
- one or more memories storing instructions;
- and
- one or more processors coupled to the one or more memories that, when executing the instructions, perform the steps of:
- accessing metadata associated with a first portion of a first downscaled video that has a first resolution and was generated using a trained downscaling convolutional neural network;
- determining, based on the metadata, a first upscaler that should be used when upscaling the first portion of the first downscaled video;
- and
- executing the first upscaler on the first portion of the first downscaled video to generate a first portion of a first reconstructed video that is accessible for playback and has a second resolution that is greater than the first resolution.
Dependencies are inferred from the claim text.