Enhanced architecture for deep learning-based video processing

A device comprising:

• a first neural network to filter images of a video sequence;
• a second neural network to generate kernel weights for the first neural network, the second neural network having a different architecture than the first neural network, the first neural network and the second neural network to operate on different images of the video sequence in parallel;
• machine-readable instructions;
• and
• at least one processor circuit to be programmed based on the machine-readable instructions to:
• cause the first neural network to generate filtered image data based on a first image of the video sequence and first kernel weights from the second neural network;
• and
• cause the second neural network to operate in parallel with the first neural network to generate second kernel weights for the first neural network based on a second image of the video sequence, the second image subsequent to the first image in the video sequence.