TY - GEN
T1 - Fully Neural Network Mode Based Intra Prediction of Variable Block Size
AU - Sun, Heming
AU - Yu, Lu
AU - Katto, Jiro
N1 - Publisher Copyright:
© 2020 IEEE.
Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.
PY - 2020/12/1
Y1 - 2020/12/1
N2 - Intra prediction is an essential component in the image coding. This paper gives an intra prediction framework completely based on neural network modes (NM). Each NM can be regarded as a regression from the neighboring reference blocks to the current coding block. (1) For variable block size, we utilize different network structures. For small blocks 4×4 and 8×8, fully connected networks are used, while for large blocks 16×16 and 32×32, convolutional neural networks are exploited. (2) For each prediction mode, we develop a specific pre-trained network to boost the regression accuracy. When integrating into HEVC test model, we can save 3.55%, 3.03% and 3.27% BD-rate for Y, U, V components compared with the anchor. As far as we know, this is the first work to explore a fully NM based framework for intra prediction, and we reach a better coding gain with a lower complexity compared with the previous work.
AB - Intra prediction is an essential component in the image coding. This paper gives an intra prediction framework completely based on neural network modes (NM). Each NM can be regarded as a regression from the neighboring reference blocks to the current coding block. (1) For variable block size, we utilize different network structures. For small blocks 4×4 and 8×8, fully connected networks are used, while for large blocks 16×16 and 32×32, convolutional neural networks are exploited. (2) For each prediction mode, we develop a specific pre-trained network to boost the regression accuracy. When integrating into HEVC test model, we can save 3.55%, 3.03% and 3.27% BD-rate for Y, U, V components compared with the anchor. As far as we know, this is the first work to explore a fully NM based framework for intra prediction, and we reach a better coding gain with a lower complexity compared with the previous work.
KW - convolutional neural network
KW - deep learning
KW - fully connected layer
KW - image compression
KW - Intra prediction
UR - http://www.scopus.com/inward/record.url?scp=85099456147&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85099456147&partnerID=8YFLogxK
U2 - 10.1109/VCIP49819.2020.9301842
DO - 10.1109/VCIP49819.2020.9301842
M3 - Conference contribution
AN - SCOPUS:85099456147
T3 - 2020 IEEE International Conference on Visual Communications and Image Processing, VCIP 2020
SP - 21
EP - 24
BT - 2020 IEEE International Conference on Visual Communications and Image Processing, VCIP 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2020 IEEE International Conference on Visual Communications and Image Processing, VCIP 2020
Y2 - 1 December 2020 through 4 December 2020
ER -