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
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.
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.

Alle Ressourcen in diesem Repository sind urheberrechtlich geschützt, soweit nicht anderweitig angezeigt.