Details
- Affiliation
-
AffiliationTechnische Universität München
- Country
-
CountryGermany
Lossy compressed images usually suffer from unpleasant artifacts, especially when the bitrate is low. In order to improve the image quality while keeping the bit rate the same, decoder-side compression artifacts reduction (CAR) becomes necessary. Recently, convolutional neural networks are adopted for CAR tasks and achieve the state-of-the-art performance. However, most CAR algorithms only focus on the reconstruction of the luminance channel. Also, they train a separate model for each quality factor (QF), which makes these approaches not practical in a real codec. In this paper, we propose a quality-blind training strategy to train a single model to enhance color images compressed with a wide range of quality levels. Our experimental results for three representative CAR algorithms show the superiority of the proposed quality-blind training compared to separate training. The results for pseudo and real quality-blind CAR tests further prove the generalizability of the quality-blind training for practical CAR tasks.