Reference-free system for determining quality of video data

A system comprising:

• one or more non-transitory memories storing computer-executable instructions;
• and
• one or more hardware processors to execute the computer-executable instructions to:
• provide first video data to a first device;
• receive first user input indicative of a first score, wherein the first score is indicative of a quality of the first video data;
• determine one or more first characteristics of the first video data;
• use the first score and the one or more first characteristics of the first video data as training data to train a machine learning system to determine scores based on characteristics of video data;
• determine one or more second characteristics of a first frame of second video data;
• process the second video data using the machine learning system to determine a second score associated with the first frame based on the one or more second characteristics;
• determine one or more of:
• a network performance characteristic, a device characteristic, or an application performance characteristic associated with the first frame;
• and
• generate an output that indicates the first frame and the one or more of the network performance characteristic, the device characteristic, or the application performance characteristic.

A system comprising:

• one or more non-transitory memories storing computer-executable instructions;
• and
• one or more hardware processors to execute the computer-executable instructions to:
• provide first data to one or more first devices associated with a first user, wherein the first data is associated with a first score indicative of a quality of the first data;
• receive first user input indicating a second score indicative of the quality of the first data;
• determine a first difference between the first score and the second score;
• provide second data to the one or more first devices, wherein the second data is associated with one or more first characteristics;
• receive second user input indicating a third score indicative of a quality of the second data;
• and
• train a machine learning system to determine scores based on characteristics of data using the first difference, the third score, and the one or more first characteristics as training data.