轻量级视频去噪模型，实时处理真实相机噪声。
PocketDVDNet: Realtime Video Denoising for Real Camera Noise
- 通过稀疏剪枝与物理噪声建模压缩模型，提升效率。
- 模型尺寸减小74%，仍保持高质量去噪效果。
- 适合自动驾驶、监控等实时视频应用。
真实场景下的多成分传感器噪声使实时视频去噪在自动对焦、自动驾驶和监控等应用中仍具挑战。本文提出PocketDVDNet，一种基于模型压缩框架的轻量级视频去噪器，结合稀疏引导的结构化剪枝、物理启发的噪声模型和知识蒸馏，在降低资源消耗的同时实现高质量恢复。从参考模型出发，诱导稀疏性，进行定向通道剪枝，并在真实多成分噪声上重训练教师模型。学生网络学习隐式噪声处理机制，无需显式噪声图输入。PocketDVDNet将原模型大小减少74%，同时提升去噪质量，可实时处理5帧图像块。结果表明，激进压缩与领域自适应蒸馏相结合，能有效平衡性能与效率，适用于实际实时视频去噪。
原文摘要 · Abstract (English)
Live video denoising under realistic, multi-component sensor noise remains challenging for applications such as autofocus, autonomous driving, and surveillance. We propose PocketDVDNet, a lightweight video denoiser developed using our model compression framework that combines sparsity-guided structured pruning, a physics-informed noise model, and knowledge distillation to achieve high-quality restoration with reduced resource demands. Starting from a reference model, we induce sparsity, apply targeted channel pruning, and retrain a teacher on realistic multi-component noise. The student network learns implicit noise handling, eliminating the need for explicit noise-map inputs. PocketDVDNet reduces the original model size by 74% while improving denoising quality and processing 5-frame patches in real-time. These results demonstrate that aggressive compression, combined with domain-adapted distillation, can reconcile performance and efficiency for practical, real-time video denoising.
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。