用图注意力机制减少冗余特征，提升图像压缩效率。
GABIC: Graph-based Attention Block for Image Compression
- 基于k近邻增强的图注意力机制，抑制特征冗余。
- 高比特率下优于同类方法，压缩性能显著提升。
- 适合关注高效图像压缩的算法研发者。
尽管JPEG和HEVC-intra等标准化编码器仍是图像压缩的行业标准，神经网络学习图像压缩（LIC）编码器代表了有前景的替代方案。具体而言，将视觉变换器中的注意力机制引入LIC模型已显示出更高的压缩效率。然而，额外的效率往往伴随特征冗余的聚集。本文提出图基注意力块（GABIC），一种基于k-近邻增强注意力机制以减少特征冗余的方法。实验表明，GABIC在高比特率下优于对比方法，显著提升了压缩性能。
原文摘要 · Abstract (English)
While standardized codecs like JPEG and HEVC-intra represent the industry standard in image compression, neural Learned Image Compression (LIC) codecs represent a promising alternative. In detail, integrating attention mechanisms from Vision Transformers into LIC models has shown improved compression efficiency. However, extra efficiency often comes at the cost of aggregating redundant features. This work proposes a Graph-based Attention Block for Image Compression (GABIC), a method to reduce feature redundancy based on a k-Nearest Neighbors enhanced attention mechanism. Our experiments show that GABIC outperforms comparable methods, particularly at high bit rates, enhancing compression performance.
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。