TY - GEN
T1 - Layered Image Compression Using Scalable Auto-Encoder
AU - Jia, Chuanmin
AU - Liu, Zhaoyi
AU - Wang, Yao
AU - Ma, Siwei
AU - Gao, Wen
N1 - Funding Information:
ACKNOWLEDGMENT The authors would like to thank Dr. Johannes Ballé for kindly providing their trained models of [11] and the decoded images of [13] for performance comparison. This work was done by C. Jia and Z. Liu as visiting students in NYU-Tandon sponsored by China Scholarship Council (CSC), which is gratefully acknowledged.
Publisher Copyright:
© 2019 IEEE.
PY - 2019/4/22
Y1 - 2019/4/22
N2 - This paper presents a novel convolutional neural network (CNN) based image compression framework via scalable auto-encoder (SAE). Specifically, our SAE based deep image codec consists of hierarchical coding layers, each of which is an end-to-end optimized auto-encoder. The coarse image content and texture are encoded through the first (base) layer while the consecutive (enhance) layers iteratively code the pixel-level reconstruction errors between the original and former reconstructed images. The proposed SAE structure alleviates the need to train multiple models for different bit-rate points by recently proposed auto-encoder based codecs. The SAE layers can be combined to realize multiple rate points, or to produce a scalable stream. The proposed method has similar rate-distortion performance in the low-to-medium rate range as the state-of-the-art CNN based image codec (which uses different optimized networks to realize different bit rates) over a standard public image dataset. Furthermore, the proposed codec generates better perceptual quality in this bit rate range.
AB - This paper presents a novel convolutional neural network (CNN) based image compression framework via scalable auto-encoder (SAE). Specifically, our SAE based deep image codec consists of hierarchical coding layers, each of which is an end-to-end optimized auto-encoder. The coarse image content and texture are encoded through the first (base) layer while the consecutive (enhance) layers iteratively code the pixel-level reconstruction errors between the original and former reconstructed images. The proposed SAE structure alleviates the need to train multiple models for different bit-rate points by recently proposed auto-encoder based codecs. The SAE layers can be combined to realize multiple rate points, or to produce a scalable stream. The proposed method has similar rate-distortion performance in the low-to-medium rate range as the state-of-the-art CNN based image codec (which uses different optimized networks to realize different bit rates) over a standard public image dataset. Furthermore, the proposed codec generates better perceptual quality in this bit rate range.
KW - CNN
KW - Image Compression
KW - end-to-end optimization
KW - scalable auto-encoder
UR - http://www.scopus.com/inward/record.url?scp=85065605547&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85065605547&partnerID=8YFLogxK
U2 - 10.1109/MIPR.2019.00087
DO - 10.1109/MIPR.2019.00087
M3 - Conference contribution
AN - SCOPUS:85065605547
T3 - Proceedings - 2nd International Conference on Multimedia Information Processing and Retrieval, MIPR 2019
SP - 431
EP - 436
BT - Proceedings - 2nd International Conference on Multimedia Information Processing and Retrieval, MIPR 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2nd International Conference on Multimedia Information Processing and Retrieval, MIPR 2019
Y2 - 28 March 2019 through 30 March 2019
ER -