| Publikationstyp: | Konferenz: Paper |
| Art der Begutachtung: | Peer review (Publikation) |
| Titel: | Observer dependent lossy image compression |
| Autor/-in: | Weber, Maurice Renggli, Cedric Grabner, Helmut Zhang, Ce |
| et. al: | No |
| DOI: | 10.48550/arXiv.1910.03472 |
| Angaben zur Konferenz: | 42nd German Conference on Pattern Recognition (DAGM-GCPR), virtual, 28 September - 1 October 2020 |
| Erscheinungsdatum: | Okt-2020 |
| Sprache: | Englisch |
| Fachgebiet (DDC): | 006: Spezielle Computerverfahren |
| Zusammenfassung: | Deep neural networks have recently advanced the state-of-the-art in image compression and surpassed many traditional compression algorithms. The training of such networks involves carefully trading off entropy of the latent representation against reconstruction quality. The term quality crucially depends on the observer of the images which, in the vast majority of literature, is assumed to be human. In this paper, we aim to go beyond this notion of compression quality and look at human visual perception and image classification simultaneously. To that end, we use a family of loss functions that allows to optimize deep image compression depending on the observer and to interpolate between human perceived visual quality and classification accuracy, enabling a more unified view on image compression. Our extensive experiments show that using perceptual loss functions to train a compression system preserves classification accuracy much better than traditional codecs such as BPG without requiring retraining of classifiers on compressed images. For example, compressing ImageNet to 0.25 bpp reduces Inception-ResNet classification accuracy by only 2%. At the same time, when using a human friendly loss function, the same compression system achieves competitive performance in terms of MS-SSIM. By combining these two objective functions, we show that there is a pronounced trade-off in compression quality between the human visual system and classification accuracy. |
| URI: | https://digitalcollection.zhaw.ch/handle/11475/21554 |
| Volltext Version: | Publizierte Version |
| Lizenz (gemäss Verlagsvertrag): | Lizenz gemäss Verlagsvertrag |
| Departement: | School of Engineering |
| Organisationseinheit: | Institut für Datenanalyse und Prozessdesign (IDP) |
| Enthalten in den Sammlungen: | Publikationen School of Engineering |
Dateien zu dieser Ressource:
Es gibt keine Dateien zu dieser Ressource.
Zur Langanzeige
Weber, M., Renggli, C., Grabner, H., & Zhang, C. (2020, October). Observer dependent lossy image compression. 42nd German Conference on Pattern Recognition (DAGM-GCPR), Virtual, 28 September - 1 October 2020. https://doi.org/10.48550/arXiv.1910.03472
Weber, M. et al. (2020) ‘Observer dependent lossy image compression’, in 42nd German Conference on Pattern Recognition (DAGM-GCPR), virtual, 28 September - 1 October 2020. Available at: https://doi.org/10.48550/arXiv.1910.03472.
M. Weber, C. Renggli, H. Grabner, and C. Zhang, “Observer dependent lossy image compression,” in 42nd German Conference on Pattern Recognition (DAGM-GCPR), virtual, 28 September - 1 October 2020, Oct. 2020. doi: 10.48550/arXiv.1910.03472.
WEBER, Maurice, Cedric RENGGLI, Helmut GRABNER und Ce ZHANG, 2020. Observer dependent lossy image compression. In: 42nd German Conference on Pattern Recognition (DAGM-GCPR), virtual, 28 September - 1 October 2020. Conference paper. Oktober 2020
Weber, Maurice, Cedric Renggli, Helmut Grabner, and Ce Zhang. 2020. “Observer Dependent Lossy Image Compression.” Conference paper. In 42nd German Conference on Pattern Recognition (DAGM-GCPR), Virtual, 28 September - 1 October 2020. https://doi.org/10.48550/arXiv.1910.03472.
Weber, Maurice, et al. “Observer Dependent Lossy Image Compression.” 42nd German Conference on Pattern Recognition (DAGM-GCPR), Virtual, 28 September - 1 October 2020, 2020, https://doi.org/10.48550/arXiv.1910.03472.
Alle Ressourcen in diesem Repository sind urheberrechtlich geschützt, soweit nicht anderweitig angezeigt.