TreeNet: A Light Weight Model for Low Bitrate Image Compression

2025-12-18

AI Summary

Learning-based image compression models suffer from high computational complexity, hindering practical deployment—especially under low-bitrate constraints. Method: This paper proposes a lightweight binary-tree-structured codec architecture tailored for low-bitrate scenarios. It introduces, for the first time, a binary-tree encoder-decoder framework that integrates attention-driven multi-branch feature fusion with end-to-end differentiable entropy coding optimization—preserving multi-scale representation capability while drastically reducing parameter count and computational cost. Contribution/Results: Experiments on standard benchmark datasets demonstrate that our method achieves an average 4.83% BD-rate improvement over JPEG AI and reduces computational complexity by 87.82%, significantly outperforming existing lightweight approaches. The core contribution lies in pioneering the incorporation of binary-tree topology into the backbone architecture of image compression, enabling synergistic optimization of high compression efficiency and ultra-low inference cost.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Learning & Optimization for CV
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Reducing computational complexity remains a critical challenge for the widespread adoption of learning-based image compression techniques. In this work, we propose TreeNet, a novel low-complexity image compression model that leverages a binary tree-structured encoder-decoder architecture to achieve efficient representation and reconstruction. We employ attentional feature fusion mechanism to effectively integrate features from multiple branches. We evaluate TreeNet on three widely used benchmark datasets and compare its performance against competing methods including JPEG AI, a recent standard in learning-based image compression. At low bitrates, TreeNet achieves an average improvement of 4.83% in BD-rate over JPEG AI, while reducing model complexity by 87.82%. Furthermore, we conduct extensive ablation studies to investigate the influence of various latent representations within TreeNet, offering deeper insights into the factors contributing to reconstruction.

Problem

• Develops a low-complexity model for image compression
• Improves compression efficiency at low bitrates over standards
• Analyzes internal representations to understand reconstruction factors

Innovation

• Binary tree-structured encoder-decoder architecture for compression
• Attentional feature fusion to integrate multiple branch features
• Achieves higher compression efficiency with significantly reduced complexity