CNN ENCODING METHOD AND APPARATUS COMPRISING CONVOLUTIONAL NEURAL NETWORKCNN BASED IN-LOOP FILTER AND DECODING METHOD AND APPARATUS COMPRISING CONVOLUTIONAL NEURAL NETWORKCNN BASED IN-LOOP FILTER

摘要

An encoding apparatus and a decoding apparatus including a CNN-based in-loop filter are disclosed. The encoding apparatus includes a filtering part for generating filtering information by filtering a residual image corresponding to a difference between an original image and a prediction image, an inverse filtering part for generating inverse filtering information by inversely filtering the filtering information, a prediction part for generating the prediction image based on the original image and reconstruction information, the CNN-based in-loop filter for receiving the inverse filtering information and the prediction image and outputting the reconstruction information; and an encoding part for performing encoding based on the filtering information and the prediction image information. Therefore, it is possible to remove block boundary artifact, ringing artifact, blurring artifact, etc. through in-loop filtering.