Artifacts reduction for very low bitrate image compression with generative adversarial networks - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

Artifacts reduction for very low bitrate image compression with generative adversarial networks

Résumé

Image compression at very low bitrate has known a new dawn since the democratization of convolutional neural networks (CNNs). A promising approach consists in using an « end-to-end » neural network-based compression scheme, performing both coding and decoding phases. Such techniques lead to perceptually convincing results, notably with GAN-based architectures. However, due to the nature of GAN, alongside being computationally expensive on the encoder side, such schemes cannot ensure the preservation of fine details, such as small digits or letters on an ID card. To overcome this limitation, we propose a new model, specifically trained to reduce artifacts resulting from strong lossy compressions. An advantage of the proposed approach comes from the fact that it can be interpreted as a post-processing step that can easily be added to any compression scheme without modifying the codec. Experimental results show that our model perceptually outperforms the state-of-the-art compression standards for very low bitrates.
Fichier non déposé

Dates et versions

hal-02454683 , version 1 (24-01-2020)

Identifiants

Citer

Sebastien Hamis, Titus Zaharia, Olivier Rousseau. Artifacts reduction for very low bitrate image compression with generative adversarial networks. ICCE-Berlin 2019: 9th International Conference on Consumer Electronics, Sep 2019, Berlin, Germany. pp.76-81, ⟨10.1109/ICCE-Berlin47944.2019.8966158⟩. ⟨hal-02454683⟩
12 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More