US 11,670,011 B2
Image compression apparatus and learning apparatus and method for the same
Sang Youn Lee, Seoul (KR); Tae Oh Kim, Seoul (KR); Han Bin Son, Seoul (KR); and Hyeong Min Lee, Seoul (KR)
Assigned to INDUSTRY-ACADEMIC COOPERATION FOUNDATION YONSEI UNIVERSITY, Seoul (KR)
Filed by INDUSTRY-ACADEMIC COOPERATION FOUNDATION, YONSEI UNIVERSITY, Seoul (KR)
Filed on Jan. 11, 2021, as Appl. No. 17/146,313.
Prior Publication US 2022/0222864 A1, Jul. 14, 2022
Int. Cl. G06T 9/00 (2006.01); G06T 3/40 (2006.01); G06N 20/20 (2019.01); G06N 3/084 (2023.01); G06N 3/045 (2023.01)
CPC G06T 9/002 (2013.01) [G06N 3/045 (2023.01); G06N 3/084 (2013.01); G06N 20/20 (2019.01); G06T 3/4046 (2013.01)] 15 Claims
OG exemplary drawing
 
1. An image compression apparatus comprising:
an image acquisition unit configured to acquire a raw data image;
a pre-processing network configured to receive the raw data image and pre-process the raw data image according to a pattern estimation method learned beforehand; and
an encoder unit configured to receive the pre-processed image and compress the pre-processed image according to a pre-designated standard compression technique to output a compressed image,
wherein the pre-processing network is added during learning and is implemented as an artificial neural network, and the pre-processing network has learned beforehand by way of a backpropagation of a restoration error through a codec modeling unit, the codec modeling unit having learned beforehand to simulate a standard codec unit comprising the encoder unit and a decoder unit configured to obtain a decoded image by receiving and decoding the compressed image, the restoration error obtained by comparing a restored image with the raw data image, the restored image obtained based on a simulated decoded image outputted from the codec modeling unit.