Video encoding optimization of extended spaces including last stage processes

A video coding system, comprising:

• a first format converter that converts a video signal from a first format to a second format;
• a video codec, comprising:
• an encoder that performs motion-compensated prediction on the video signal in the second format, including a subtractor that combines predicted images in the second format with the video signal in the second format;
• a decoder that decodes portions of coded video output by the encoder and outputs decoded data in the second format;
• a filtering system having a first input of the output of the decoder in the second format and a second input of the video signal in the first format, and producing a filtered output in the second format using both the first input and the second input;
• a decoded picture buffer to store decoded pictures output by the filtering system;
• and
• a prediction system that selects parameters of the motion-compensated prediction.

A non-transitory computer readable medium storing program instructions that, when executed by a processing device, causes the device to:

• convert a video signal from a first format to a second format;
• code the converted video signal by motion-compensated prediction, the coding including combining predicted images in the second format with the video signal in the second format;
• decode portions of the coded video to produce decoded data in the second format;
• filter with a first input of the decoded data in the second format and a second input of the video signal in the first format, and producing filtered decoded data in the second format using both the first input and the second input;
• store filtered picture data in a decoded picture buffer;
• and
• convert the decoded portions from the second format to the first format.