TM-BSN：用于真实世界自监督图像去噪的三角掩码盲点网络
TM-BSN: Triangular-Masked Blind-Spot Network for Real-World Self-Supervised Image Denoising
提出三角掩码卷积，在原始分辨率下构造菱形盲点以匹配去马赛克噪声的相关性几何，避免下采样带来的噪声统计改变和上下文损失；并用知识蒸馏将多个盲点预测集成到轻量U-Net，为真实图像自监督去噪提供了可复用的架构设计。
原文依据 we introduce a triangular-masked convolution that restricts the kernel to its upper-triangular region, creating a diamond-shaped blind spot at the original resolution.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论盲点网络（BSN）通过阻止访问目标像素来实现自监督图像去噪，无需真实标签即可估计干净信号。然而，该方法假设像素级噪声独立，而真实sRGB图像由于相机图像信号处理（ISP）流程产生的空间相关噪声而违背了这一假设。尽管一些方法采用下采样来去相关噪声，但它们改变了噪声统计特性并限制了网络利用完整上下文信息的能力。本文提出三角掩码盲点网络（TM-BSN），一种新颖的盲点架构，能够准确建模真实sRGB噪声的空间相关性。这种相关性源于去马赛克过程，其中每个像素由邻域样本以空间衰减权重重建，形成菱形模式。为使感受野与该几何结构对齐，我们引入三角掩码卷积，将卷积核限制在上三角区域，在原始分辨率下创建菱形盲点。该设计排除了相关像素，同时充分利用不相关上下文，无需下采样或后处理。此外，我们使用知识蒸馏将多个盲点预测的互补知识迁移到轻量U-Net中，同时提升准确性和效率。在真实世界基准上的大量实验表明，我们的方法达到了最优性能，显著优于现有自监督方法。代码见 https://github.com/parkjun210/TM-BSN。
英文原文摘要 1672 字符
Blind-spot networks (BSNs) enable self-supervised image denoising by preventing access to the target pixel, allowing clean signal estimation without ground-truth supervision. However, this approach assumes pixel-wise noise independence, which is violated in real-world sRGB images due to spatially correlated noise from the camera's image signal processing (ISP) pipeline. While several methods employ downsampling to decorrelate noise, they alter noise statistics and limit the network's ability to utilize full contextual information. In this paper, we propose the Triangular-Masked Blind-Spot Network (TM-BSN), a novel blind-spot architecture that accurately models the spatial correlation of real sRGB noise. This correlation originates from demosaicing, where each pixel is reconstructed from neighboring samples with spatially decaying weights, resulting in a diamond-shaped pattern. To align the receptive field with this geometry, we introduce a triangular-masked convolution that restricts the kernel to its upper-triangular region, creating a diamond-shaped blind spot at the original resolution. This design excludes correlated pixels while fully leveraging uncorrelated context, eliminating the need for downsampling or post-processing. Furthermore, we use knowledge distillation to transfer complementary knowledge from multiple blind-spot predictions into a lightweight U-Net, improving both accuracy and efficiency. Extensive experiments on real-world benchmarks demonstrate that our method achieves state-of-the-art performance, significantly outperforming existing self-supervised approaches. Our code is available at https://github.com/parkjun210/TM-BSN.