Input filtering in a video encoder
View Patent ↗Presented herein are systems, methods, and apparatus for real-time high definition television encoding. In one embodiment, there is a method for encoding video data. The method comprises performing motion estimation in parallel of a plurality of pictures using original reference pictures associated with the plurality of pictures; examining the plurality of pictures to determine whether their content is complex; and low pass filtering pictures with complex content to smooth them before proceeding to continue with the encoding process.
1. A method for encoding video data, the method comprising:
estimating amounts of data for encoding a plurality of pictures in parallel using original reference pictures associated with the plurality of pictures with a plurality of encoders, wherein the estimating includes lossy encoding of the plurality of pictures using reconstructed reference pictures using a master encoder and providing target rates to the plurality of encoders using the master encoder, further wherein the original reference pictures are unaltered by the plurality of encoders;
classifying the plurality of pictures according to their complexity;
low pass filtering pictures from the plurality of pictures with high complexity.
2. The method according to claim 1 wherein the complexity indicates an amount of high frequency content associated with the plurality of pictures.
3. The method of claim 1 wherein the original reference pictures are scaled down.
4. A system for encoding video data, the system comprising:
a hardware accelerator for estimating amounts of data for encoding a plurality of pictures in parallel using original reference pictures associated with the plurality of pictures, wherein the original reference pictures are unaltered by the hardware accelerator;
a first engine for classifying the plurality of pictures according to their complexity;
a second engine for low pass filtering pictures from the plurality of pictures with high complexity; and
a master encoder for lossy encoding the plurality of pictures using reconstructed reference pictures, said master encoder being discrete from the plurality of encoders and providing target rates to the plurality of encoders.
5. The system according to claim 4 wherein the complexity indicates an amount of high frequency content associated with the plurality of pictures.
6. The system of claim 4 wherein the original reference pictures are scaled down.
7. A non-transitory computer readable medium having stored thereon, a computer program having at least one code section encoding video data, the at least one code section being executable by a machine for causing the machine to perform steps comprising:
estimating amounts of data for encoding a plurality of pictures in parallel using original reference pictures associated with the plurality of pictures, wherein the estimating includes lossy encoding of the plurality of pictures using reconstructed reference pictures using a master encoder and providing target rates to the plurality of encoders using the master encoder, further wherein the original reference pictures are unaltered by the computer program;
classifying the plurality of pictures according to their complexity;
low pass filtering pictures from the plurality of pictures with high complexity.
8. The non-transitory computer readable medium according to claim 7 wherein the complexity indicates an amount of high frequency content associated with the plurality of pictures.
9. A non-transitory computer readable medium of claim 7 wherein the original reference pictures are scaled down.
10. A system for encoding video data, the system comprising:
a hardware accelerator for estimating amounts of data for encoding a plurality of pictures in parallel using original reference pictures associated with the plurality of pictures, wherein the estimating includes lossy encoding of the plurality of pictures using reconstructed reference pictures using a master encoder and providing target rates to the plurality of encoders using the master encoder, further wherein the original reference pictures are unaltered by the hardware accelerator;
a circuit for classifying the plurality of pictures according to their complexity;
a low pass filter for low pass filtering pictures from the plurality of pictures with high complexity.
11. The system according to claim 10 wherein the complexity indicates an amount of high frequency content associated with the plurality of pictures.
12. The system of claim 10 wherein the original reference pictures are scaled down.
13. The method of claim 1 , wherein estimating amounts of data for encoding a plurality of pictures in parallel using original reference pictures further comprises motion estimating from reference pictures that have not lost information from quantization.
14. The system of claim 4 , wherein estimating amounts of data for encoding a plurality of pictures in parallel using original reference pictures further comprises motion estimating from reference pictures that have not lost information from quantization.
15. The non-transitory computer medium of claim 7 , wherein estimating amounts of data for encoding a plurality of pictures in parallel using original reference pictures further comprises motion estimating from reference pictures that have not lost information from quantization.
16. The system of claim 10 , wherein estimating amounts of data for encoding a plurality of pictures in parallel using original reference pictures further comprises motion estimating from reference pictures that have not lost information from quantization.
17. The method of claim 1 , wherein the plurality of pictures are from the same stream.
18. The non-transitory computer readable medium of claim 7 , wherein the plurality of instructions causes encoding the plurality of pictures in series by a device that is discrete from a device encoding the plurality of pictures in parallel.
19. The system of claim 10 , wherein the low pass filter selectively low pass filters pictures from the plurality of pictures with high complexity.