单步扩散压缩图像，速度提升46倍且保质。
One-Step Diffusion for Perceptual Image Compression
- 仅用一步扩散过程解码，大幅降低计算开销。
- 采用特征空间判别器，提升重构图像的纹理与结构质量。
- 适合对实时性要求高的图像压缩场景。
基于扩散模型的图像压缩方法已取得显著进展，在低比特率下实现高感知质量。然而，其实际应用受限于推理延迟大、计算开销高，主要源于解码时需大量去噪步骤。为此，我们提出一种仅需单步扩散过程的压缩方法，显著提升推理速度。为增强重建图像的感知质量，引入在紧凑特征表示上运行的判别器，利用特征更有效捕捉高层纹理与结构细节。实验表明，该方法在压缩性能相当的前提下，相比近期扩散模型方法实现46倍的推理加速。代码与模型已开源：https://github.com/cheesejiang/OSDiff。
原文摘要 · Abstract (English)
Diffusion-based image compression methods have achieved notable progress, delivering high perceptual quality at low bitrates. However, their practical deployment is hindered by significant inference latency and heavy computational overhead, primarily due to the large number of denoising steps required during decoding. To address this problem, we propose a diffusion-based image compression method that requires only a single-step diffusion process, significantly improving inference speed. To enhance the perceptual quality of reconstructed images, we introduce a discriminator that operates on compact feature representations instead of raw pixels, leveraging the fact that features better capture high-level texture and structural details. Experimental results show that our method delivers comparable compression performance while offering a 46$\times$ faster inference speed compared to recent diffusion-based approaches. The source code and models are available at https://github.com/cheesejiang/OSDiff.
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。