Quality-metric-agnostic rate control
A system includes a processing device to receive a video content, a quality metric, and a target bit rate for encoding the video content. The system includes encoding hardware to perform frame encoding on the video content and a controller coupled between the processing device and the encoding hardware. The controller is programmed with machine instructions to generate first QP values on a per-frame basis using a frame machine learning model with a first plurality of weights. The first plurality of weights depends at least in part on the quality metric and the target bit rate. The controller is further programmed to provide the first QP values to the encoding hardware for rate control of the frame encoding.
1. A system comprising:
a processing device to receive a video content, a quality metric, and a target bit rate for encoding the video content;
encoding hardware to perform frame encoding on the video content; and
a controller coupled between the processing device and the encoding hardware, the controller programmed with machine instructions to:
generate first quantization parameter (QP) values on a per-frame basis using a frame machine learning model with a first plurality of weights, wherein the first plurality of weights depends at least in part on the quality metric and the target bit rate; and
provide the first QP values directly to the encoding hardware for rate control of the frame encoding; and
wherein the encoding hardware is further to encode each respective frame using a respective one of the first OP values received directly from the controller.
2. The system of claim 1 , wherein the quality metric is a video quality assessment metric and the target bit rate is one of a predefined bit rate, a constant bit rate, or a variable bit rate.
3. The system of claim 1 , wherein the processing device is further to:
retrieve a plurality of parameters related to the quality metric; and
provide the plurality of parameters to the controller; and
wherein the machine instructions are further to select the first plurality of weights corresponding to the plurality of parameters.
4. The system of claim 1 , wherein the frame machine learning model is a frame reinforcement learning model that is instantiated in a neural network, wherein the neural network uses the first plurality of weights to maximize a reward function of the neural network while encoding a plurality of frames of the video content.
5. The system of claim 1 , wherein the encoding hardware is further to perform sub-frame encoding, and the machine instructions are further to:
generate second QP values on a per-sub-frame basis using a sub-frame machine learning model with a second plurality of weights, wherein the second plurality of weights depends at least in part on the quality metric and the target bit rate; and
provide the second QP values to the encoding hardware for use in rate control when encoding respective sub-frames of the video content.
6. The system of claim 5 , wherein the encoding hardware is further to encode each respective sub-frame using a respective one of the second QP values.
7. An integrated circuit comprising:
encoding hardware to perform frame encoding on video content; and
a processing device coupled to encoding hardware, wherein the processing device is to implement, using program code, a frame machine learning rate controller that is to:
receive the video content, a quality metric, and a target bit rate for encoding the video content;
generate first quantization parameter (QP) values on a per-frame basis using a frame machine learning model with a first plurality of weights, wherein the first plurality of weights depends at least in part on the quality metric and the target bit rate; and
provide the first QP values directly to the encoding hardware for rate control of the frame encoding; and
wherein the encoding hardware is further to encode each respective frame using a respective one of the first QP values received directly from the processing device.
8. The integrated circuit of claim 7 , wherein the quality metric is a video quality assessment metric and the target bit rate is one of a predefined bit rate, a constant bit rate, or a variable bit rate.
9. The integrated circuit of claim 7 , wherein the processing device is further to:
retrieve a plurality of parameters related to the quality metric; and
select the first plurality of weights corresponding to the plurality of parameters.
10. The integrated circuit of claim 7 , wherein the frame machine learning model is a frame reinforcement learning model that is instantiated in a neural network, wherein the neural network uses the first plurality of weights to maximize a reward function of the neural network while encoding a plurality of frames of the video content.
11. The integrated circuit of claim 7 , wherein the encoding hardware is further to perform sub-frame encoding, further comprising a controller coupled to the processing device and to execute machine instructions to:
generate second QP values on a per-sub-frame basis using a sub-frame machine learning model with a second plurality of weights, wherein the second plurality of weights depends at least in part on the quality metric and the target bit rate; and
provide the second QP values to the encoding hardware for use in rate control when encoding respective sub-frames of the video content.
12. The integrated circuit of claim 11 , wherein the encoding hardware is further to encode each respective sub-frame using a respective one of the second QP values.
13. A method comprising:
receiving video content, a quality metric, and a target bit rate for encoding the video content;
generating, by a processing device executing program code, first quantization parameter (QP) values on a per-frame basis using a frame machine learning model with a first plurality of weights, wherein the first plurality of weights depends at least in part on the quality metric and the target bit rate;
providing by the processing device, the first QP values directly to encoding hardware for rate control of frame encoding by the encoding hardware; and
encoding, with the encoding hardware, each respective frame using a respective one of the first QP values received directly from the processing device.
14. The method of claim 13 , wherein the quality metric is a video quality assessment metric and the target bit rate is one of a predefined bit rate, a constant bit rate, or a variable bit rate.
15. The method of claim 13 , further comprising:
retrieving a plurality of parameters related to the quality metric; and
selecting the first plurality of weights corresponding to the plurality of parameters.
16. The method of claim 13 , wherein the frame machine learning model is a frame reinforcement learning model that is instantiated in a neural network, wherein the neural network uses the first plurality of weights to maximize a reward function of the neural network while encoding a plurality of frames of the video content.
17. The method of claim 13 , wherein the encoding hardware is further to perform sub-frame encoding, the method further comprising:
generating second QP values on a per-sub-frame basis using a sub-frame machine learning model with a second plurality of weights, wherein the second plurality of weights depends at least in part on the quality metric and the target bit rate; and
providing the second QP values to the encoding hardware for use in rate control when encoding respective sub-frames of the video content.
18. The method of claim 17 , further comprising encoding, with the encoding hardware, each respective sub-frame using a respective one of the second QP values.