Streaming system and method
AssigneeSony Interactive Entertainment Inc.
InventorsMatthew William Sanders, Justin Lim, Sahin Serdar Kocdemir, Marina Villanueva Barreiro
Life of this patent
Estimating the term…
Abstract
A method of displaying streamed content includes receiving streamed content from a server, decoding the streamed content, outputting the decoded streamed content for display, and estimating a degree of attention of a user on the streamed content, where the decoding step comprises decoding the content to a quality that is dependent upon the estimated degree of attention of the use.
Claims11
A client device for displaying streamed content, comprising:
- a receiver to receive streamed content from a server;
- a variable decoder configured to decode the streamed content in a decoding process, with successive stages of the decoding process decoding the streamed content to a higher quality from decoded outputs of prior stages;
- an output for displaying the decoded streamed content;
- and
- an attention estimation processor configured to estimate a degree of attention of a user on the streamed content;
- wherein
- the streamed content has been encoded using an encoding process with a trained multi-stage neural network autoencoder configured for preserving size and spatial resolution of the streamed content in a plurality of convolution layers,
- the variable decoder is further configured to decode the streamed content by performing at least one stage of the decoding process in a reversal of the encoding process with the trained multi-stage neural network autoencoder based only on the streamed content, and
- the variable decoder is configured to selectively decode the streamed content in accordance with the estimated degree of attention of the user.
A method of displaying streamed content at a client device, comprising:
- receiving streamed content from a server;
- decoding the streamed content using a variable decoder in a decoding process, wherein successive stages of the decoding process decode the streamed content to a higher quality from decoded outputs of prior stages;
- displaying the decoded streamed content;
- and
- estimating a degree of attention of a user on the streamed content;
- wherein
- the streamed content has been encoded using an encoding process with a trained multi-stage neural network autoencoder configured for preserving size and spatial resolution of the streamed content in a plurality of convolution layers,
- decoding includes decoding the streamed content by performing at least one stage of the decoding process in a reversal of the encoding process with the trained multi-stage neural network autoencoder based only on the streamed content, and
- decoding further comprises selectively decoding the streamed content in accordance with the estimated degree of attention of the user.
A non-transitory, computer readable storage medium containing computer program comprising computer executable instructions adapted to cause a computer system to perform operations comprising:
- receiving streamed content from a server;
- decoding the streamed content using a variable decoder in a decoding process, wherein successive stages of the decoding process decode the streamed content to a higher quality from decoded outputs of prior stages;
- displaying the decoded streamed content;
- and
- estimating a degree of attention of a user on the streamed content;
- wherein
- the streamed content has been encoded using an encoding process with a trained multi-stage neural network autoencoder configured for preserving size and spatial resolution of the streamed content in a plurality of convolution layers,
- decoding includes decoding the streamed content by performing at least one stage of the decoding process in a reversal of the encoding process with the trained multi-stage neural network autoencoder based only on the streamed content, and
- decoding further comprises selectively decoding the streamed content in accordance with the estimated degree of attention of the user.
Dependencies are inferred from the claim text.