轻量级神经预滤波器提升图像感知压缩效率
JND-Guided Light-Weight Neural Pre-Filter for Perceptual Image Coding
- 基于频域JND设计轻量CNN，实现高效感知预处理
- 1080p图像仅需7.15 GFLOPs，为同类模型的14.1%
- 开源统一平台支持可复现研究，适合压缩算法开发者
感知图像编码中，基于人眼可察觉失真（JND）的预滤波技术能有效提升压缩效率。然而现有方法计算开销大，且缺乏标准化评测基准。本文提出两项贡献：首先，构建并开源了FJNDF-Pytorch，一个面向频域JND预滤波的统一评测平台；其次，基于该平台，设计了一种全新的轻量级卷积神经网络学习框架。实验表明，所提方法在多个数据集与编码器上均达到当前最优压缩性能。在计算成本方面，处理1080p图像仅需7.15 GFLOPs，仅为近期轻量网络的14.1%。本工作在性能与效率上均表现卓越，并提供可复现的研究平台。代码已开源：https://github.com/viplab-fudan/FJNDF-Pytorch。
原文摘要 · Abstract (English)
Just Noticeable Distortion (JND)-guided pre-filter is a promising technique for improving the perceptual compression efficiency of image coding. However, existing methods are often computationally expensive, and the field lacks standardized benchmarks for fair comparison. To address these challenges, this paper introduces a twofold contribution. First, we develop and open-source FJNDF-Pytorch, a unified benchmark for frequency-domain JND-Guided pre-filters. Second, leveraging this platform, we propose a complete learning framework for a novel, lightweight Convolutional Neural Network (CNN). Experimental results demonstrate that our proposed method achieves state-of-the-art compression efficiency, consistently outperforming competitors across multiple datasets and encoders. In terms of computational cost, our model is exceptionally lightweight, requiring only 7.15 GFLOPs to process a 1080p image, which is merely 14.1% of the cost of recent lightweight network. Our work presents a robust, state-of-the-art solution that excels in both performance and efficiency, supported by a reproducible research platform. The open-source implementation is available at https://github.com/viplab-fudan/FJNDF-Pytorch.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。