Stream quality enhancement

A system comprising:

• one or more processors;
• and
• one or more computer-readable storage media collectively storing program instructions which, when executed by the one or more processors, are configured to cause the one or more processors to perform a method comprising:
• determining that quality of an audio/video (A/V) stream being streamed to a first device falls below a quality threshold;
• identifying a plurality of edge processing devices in an environment of the first device;
• determining available computing resources of each of the edge processing devices;
• selecting at least one edge processing device of the plurality of edge processing devices with sufficient computing resources for correcting quality of the A/V stream using a generative adversarial network (GAN) model;
• instructing the selected at least one edge processing device to correct the A/V stream using the GAN model to generate a corrected A/V stream, wherein generating the corrected A/V stream further comprises:
• generating, by the GAN model, high resolution frames for the corrected A/V stream using low resolution frames of the A/V stream as inputs, wherein the corrected A/V stream satisfies the quality threshold;
• and
• issuing a command to transmit the corrected A/V stream to the first device.

A computer program product comprising one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions comprising instructions configured to cause one or more processors to perform a method comprising:

• determining that quality of an audio/video (A/V) stream being streamed to a first device falls below a quality threshold;
• identifying a plurality of edge processing devices in an environment of the first device;
• determining available computing resources of each of the edge processing devices;
• selecting at least one edge processing device of the plurality of edge processing devices with sufficient computing resources for correcting quality of the A/V stream using a generative adversarial network (GAN) model;
• instructing the selected at least one edge processing device to correct the A/V stream using the GAN model to generate a corrected A/V stream, wherein generating the corrected A/V stream further comprises:
• generating, by the GAN model, high resolution frames for the corrected A/V stream using low resolution frames of the A/V stream as inputs, wherein the corrected A/V stream satisfies the quality threshold;
• and
• issuing a command to transmit the corrected A/V stream to the first device.