US 12,437,365 B2
Adaptive real time discrete cosine transform image and video processing with convolutional neural network architecture
Zhu Li, Overland Park, KS (US); and Paras Maharjan, Kansas City, MO (US)
Assigned to ATOMBEAM TECHNOLOGIES INC., Moraga, CA (US)
Filed by AtomBeam Technologies Inc., Moraga, CA (US)
Filed on Nov. 27, 2024, as Appl. No. 18/961,463.
Application 18/961,463 is a continuation in part of application No. 18/596,677, filed on Mar. 6, 2024, granted, now 12,198,304.
Prior Publication US 2025/0285225 A1, Sep. 11, 2025
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 5/10 (2006.01); G06T 5/50 (2006.01); G06T 5/60 (2024.01); G06T 5/73 (2024.01)
CPC G06T 5/10 (2013.01) [G06T 5/50 (2013.01); G06T 5/60 (2024.01); G06T 5/73 (2024.01); G06T 2207/20004 (2013.01); G06T 2207/20052 (2013.01); G06T 2207/20064 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A system for adaptive real time discrete cosine transform image and video processing with convolutional neural network architecture, comprising:
a computing device comprising at least a memory and a processor;
a Discrete Cosine Transform (DCT) Block configured to receive an initial input and output a plurality subband images;
a plurality of DCT Deblur Network channels comprising a DCT Deblur Network DC channel and a DCT Deblur Network AC channels, wherein each channel is configured to process a specific frequency component;
a parallel processing unit configured to simultaneously process all DCT Deblur Network channels; and
a plurality of programming instructions stored in the memory and operable on the processor, wherein the first plurality of programming instructions, when operating on the processor, cause the computing device to:
send a degraded input through the DCT Block where the degraded image is transformed into a plurality of transformed images; and
simultaneously process the plurality of transformed images with all DCT Deblur Network channels using the parallel processing unit.