HiDE: Hierarchical Dictionary-Based Entropy Modeling for Learned Image Compression

2026-03-06

AI Summary

This work addresses the limitations of existing learned image compression methods, which struggle to effectively leverage external priors from large-scale data and are constrained by single-layer dictionary structures that hinder representational capacity. To overcome these challenges, we propose HiDE, a novel framework featuring a hierarchical dictionary architecture that separately models global structural and local textural external priors. HiDE integrates a cascaded retrieval mechanism with a context-aware, multi-receptive-field parallel parameter estimation network, substantially enhancing the accuracy of conditional probability estimation and the expressiveness of the entropy model. Experimental results demonstrate that HiDE achieves significant BD-rate savings of 18.5%, 21.99%, and 24.01% on the Kodak, CLIC, and Tecnick datasets, respectively, markedly outperforming the VTM-12.1 anchor.

Technology Category

• Computer Vision: Representation Learning for Vision
• Machine Learning: Learning on the Edge & Model Compression
• Data Mining & Knowledge Management: Data Compression

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Learned image compression (LIC) has achieved remarkable coding efficiency, where entropy modeling plays a pivotal role in minimizing bitrate through informative priors. Existing methods predominantly exploit internal contexts within the input image, yet the rich external priors embedded in large-scale training data remain largely underutilized. Recent advances in dictionary-based entropy models have demonstrated that incorporating external priors can substantially enhance compression performance. However, current approaches organize heterogeneous external priors within a single-level dictionary, resulting in imbalanced utilization and limited representational capacity. Moreover, effective entropy modeling requires not only expressive priors but also a parameter estimation network capable of interpreting them. To address these challenges, we propose HiDE, a Hierarchical Dictionary-based Entropy modeling framework for learned image compression. HiDE decomposes external priors into global structural and local detail dictionaries with cascaded retrieval, enabling structured and efficient utilization of external information. Moreover, a context-aware parameter estimator with parallel multi-receptive-field design is introduced to adaptively exploit heterogeneous contexts for accurate conditional probability estimation. Experimental results show that HiDE achieves 18.5%, 21.99%, and 24.01% BD-rate savings over VTM-12.1 on the Kodak, CLIC, and Tecnick datasets, respectively.

Problem

• learned image compression
• entropy modeling
• external priors
• dictionary-based
• hierarchical representation

Innovation

• Hierarchical Dictionary
• External Priors
• Entropy Modeling
• Learned Image Compression
• Context-aware Parameter Estimation