通过跳过低细节区域，实现视频超分辨率加速60%且无质量损失
SkipSR: Faster Super Resolution with Token Skipping
- 从低分辨率输入直接识别低细节区域并跳过计算
- 720p视频上端到端延迟降低60%，质量无感知下降
- 适合追求高效视频生成与修复的开发者和研究者
基于扩散模型的视频超分辨率是视频生成与修复的关键技术，但速度慢、成本高，限制了其在更高分辨率和更长视频上的应用。本文提出一种新方法：许多视频区域本质上细节较少，精细处理收益有限，而现有方法对所有像素一视同仁。为此，我们设计了SkipSR框架，直接从低分辨率输入识别出低细节区域，并完全跳过这些区域的计算，仅对需要增强的区域进行超分辨率重建。该策略在保持标准与一步扩散模型感知质量的同时显著降低计算量。在标准超分辨率基准测试中，该方法在720p视频上比以往模型快达60%，且无明显质量损失。视频演示见https://rccchoudhury.github.io/skipsr/
原文摘要 · Abstract (English)
Diffusion-based super-resolution (SR) is a key component in video generation and video restoration, but is slow and expensive, limiting scalability to higher resolutions and longer videos. Our key insight is that many regions in video are inherently low-detail and gain little from refinement, yet current methods process all pixels uniformly. To take advantage of this, we propose SkipSR, a simple framework for accelerating video SR by identifying low-detail regions directly from low-resolution input, then skipping computation on them entirely, only super-resolving the areas that require refinement. This simple yet effective strategy preserves perceptual quality in both standard and one-step diffusion SR models while significantly reducing computation. In standard SR benchmarks, our method achieves up to 60% faster end-to-end latency than prior models on 720p videos with no perceptible loss in quality. Video demos are available at https://rccchoudhury.github.io/skipsr/
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。