Skip to main content

Quality Assessment of Deep-Learning-Based Image Compression.

Giuseppe, V., Andrei, P., Hulusic, V. and Marco, C., 2018. Quality Assessment of Deep-Learning-Based Image Compression. In: IEEE 20th International Workshop on Multimedia Signal Processing, 29--31 August 2018, Vancouver, Canada.

Full text available as:

Quality Assessment of Deep-Learning-Based Image Compression.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.


Official URL:


Image compression standards rely on predictive coding, transform coding, quantization and entropy coding, in order to achieve high compression performance. Very recently, deep generative models have been used to optimize or replace some of these operations, with very promising results. However, so far no systematic and independent study of the coding performance of these algorithms has been carried out. In this paper, for the first time, we conduct a subjective evaluation of two recent deeplearning-based image compression algorithms, comparing them to JPEG 2000 and to the recent BPG image codec based on HEVC Intra. We found that compression approaches based on deep auto-encoders can achieve coding performance higher than JPEG 2000, and sometimes as good as BPG. We also show experimentally that the PSNR metric is to be avoided when evaluating the visual quality of deep-learning-based methods, as their artifacts have different characteristics from those of DCT or wavelet-based codecs. In particular, images compressed at low bitrate appear more natural than JPEG 2000 coded pictures, according to a no-reference naturalness measure. Our study indicates that deep generative models are likely to bring huge innovation into the video coding arena in the coming years.

Item Type:Conference or Workshop Item (Paper)
Group:Faculty of Science & Technology
ID Code:31184
Deposited By: Symplectic RT2
Deposited On:30 Aug 2018 11:16
Last Modified:14 Mar 2022 14:12


Downloads per month over past year

More statistics for this item...
Repository Staff Only -