Systems and methods for video coding

1 . A method of video encoding, the method comprising:

selecting, by an encoding device, a video application for an input video;

setting, by the encoding device, at least a boundary parameter;

determining, by the encoding device, at least a goal parameter;

generating, by the encoding device, a plurality of tool combinations, wherein each tool combination of the plurality of tool combinations is a subset of the full encoder toolset;

tabulating, by the encoding device, performance and power consumption measurements of each tool combination of the plurality of tool combinations during video encoding; and

selecting, by the encoding device, a tool combination of the plurality of tool combinations, as a function of the performance and power consumption measurements and the at least a goal parameter.

2 . The method of claim 1 , wherein setting the boundary parameter includes setting a maximal bitrate.

3 . The method of claim 2 , wherein determining the at least a goal parameter further comprises:

measuring, by the encoding device and for a full encoder toolset, at least an output parameter; and

determining the at least a goal parameter as a function of the at least an output parameter.

4 . The method of claim 2 , wherein selecting the tool combination further comprises selecting the tool combination as a function of a rate of the tool combination, a performance of the tool combination, and a complexity of the tool combination.

5 . The method of claim 4 , wherein function reduces encoder complexity.

6 . The method of claim 4 , wherein the function optimizes compression performance for the goal parameter.

7 . The method of claim 4 , wherein the function minimizes divergence from a set output rate.

8 . The method of claim 1 , wherein selecting the tool combination further comprises selecting the tool combination as a function of a machine-learning model.

9 . The method of claim 8 , wherein the machine-learning model further comprises a hybrid neural network.

10 . The method of claim 1 , further comprising:

determining that at least a tool combination of the plurality of tool combinations has a parameter violating the at least a boundary parameter; and

eliminating the at least a tool combination.

11 . A system for energy-efficient video encoding, the system comprising a power source, a video encoding device and a power measurement device coupled to the power source and encoding device for measuring power consumed by the encoding device, the encoding device configured to:

select a video application for an input video;

set at least a boundary parameter;

determine at least a goal parameter;

generate a plurality of tool combinations, wherein each tool combination of the plurality of tool combinations has fewer tools than the full encoder toolset;

tabulate performance measurements and power consumption measurements of each tool combination of the plurality of tool combinations; and

select a tool combination of the plurality of tool combinations, wherein the selected tool combination generates an optimal combination of performance and power consumption during video encoding.

12 . The encoding device of claim 11 , wherein setting the boundary parameter includes setting a maximal performance.

13 . The encoding device of claim 11 , wherein determining the at least a goal parameter further comprises:

measuring, by the encoding device and for a full encoder toolset, at least an output parameter; and

determining the at least a goal parameter as a function of the at least an output parameter.

14 . The encoding device of claim 11 , wherein selecting the tool combination further comprises selecting the tool combination as a function of a machine-learning model.

15 . The encoding device of claim 14 , wherein the machine-learning model further comprises a hybrid neural network.

16 . The encoding device of claim 11 , further configured to:

determine that at least a tool combination of the plurality of tool combinations has a parameter violating the at least a boundary parameter; and

eliminate the at least a tool combination.