单步扩散模型实现高速高保真图像压缩
Steering One-Step Diffusion Model with Fidelity-Rich Decoder for Fast Image Compression
- 用预训练VAE生成丰富潜在表示，单步解码替代多步去噪
- 在极低比特率下仍保持优异率失真感知性能，速度提升20倍以上
- 适合需要快速压缩且要求图像保真的实际应用场景
基于扩散的图像压缩已展现出出色的感知质量。然而，其存在两大关键缺陷：(1) 因多步采样导致解码延迟过高；(2) 过度依赖生成先验造成保真度下降。为此，我们提出SODEC，一种新型单步扩散图像压缩模型。我们认为，在图像压缩中，足够信息量的潜在表示可使多步精修变得不再必要。基于此，我们利用预训练的基于VAE的模型生成富含信息的潜在表示，并以单步解码替代迭代去噪过程。同时，引入保真度引导模块，促使输出更忠实于原始图像。此外，设计了速率退火训练策略，支持在极低比特率下的有效训练。大量实验表明，SODEC显著优于现有方法，在率失真感知性能上表现卓越。相比以往扩散压缩模型，解码速度提升超过20倍。代码已开源：https://github.com/zhengchen1999/SODEC。
原文摘要 · Abstract (English)
Diffusion-based image compression has demonstrated impressive perceptual performance. However, it suffers from two critical drawbacks: (1) excessive decoding latency due to multi-step sampling, and (2) poor fidelity resulting from over-reliance on generative priors. To address these issues, we propose SODEC, a novel single-step diffusion image compression model. We argue that in image compression, a sufficiently informative latent renders multi-step refinement unnecessary. Based on this insight, we leverage a pre-trained VAE-based model to produce latents with rich information, and replace the iterative denoising process with a single-step decoding. Meanwhile, to improve fidelity, we introduce the fidelity guidance module, encouraging output that is faithful to the original image. Furthermore, we design the rate annealing training strategy to enable effective training under extremely low bitrates. Extensive experiments show that SODEC significantly outperforms existing methods, achieving superior rate-distortion-perception performance. Moreover, compared to previous diffusion-based compression models, SODEC improves decoding speed by more than 20$\times$. Code is released at: https://github.com/zhengchen1999/SODEC.
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。