Frequency-Aware Hierarchical Image Compression for Humans and Machines

To achieve efficient compression for both human vision and machine perception, scalable coding methods have been proposed in recent years. However, existing methods do not fully eliminate the redundancy between features corresponding to different tasks, resulting in suboptimal coding performance. In...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Visual communications and image processing (Online) S. 1 - 5
Hauptverfasser: Luo, Yue, Zhang, Zixiang, Kuang, Jinhao, Yu, Li
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 08.12.2024
Schlagworte:
Codecs
Convolution
Correlation
Entropy
Image coding
image coding for machines
Image reconstruction
learned image compression
Machine vision
Multitasking
Redundancy
scalable image coding
Visual communication
ISSN:2642-9357
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:To achieve efficient compression for both human vision and machine perception, scalable coding methods have been proposed in recent years. However, existing methods do not fully eliminate the redundancy between features corresponding to different tasks, resulting in suboptimal coding performance. In this paper, we propose a frequency-aware hierarchical image compression framework designed for humans and machines. Specifically, we investigate task relationships from a frequency perspective, utilizing only HF information for machine vision tasks and leveraging both HF and LF features for image reconstruction. Besides, the residual block embedded octave convolution module is designed to enhance the information interaction between HF features and LF features. Additionally, a dual-frequency channel-wise entropy model is applied to reasonably exploit the correlation between different tasks, thereby improving multi-task performance. The experiments show that the proposed method offers -69.3%∼-75.3% coding gains on machine vision tasks compared to the relevant benchmarks, and -19.1% gains over state-of-the-art scalable image codec in terms of image reconstruction quality.
ISSN:2642-9357
DOI:10.1109/VCIP63160.2024.10849897