Unbiased Monte Carlo path tracing that is extensively used in realistic rendering produces undesirable noise, especially with low samples per pixel (spp). Recently, several methods have coped with this problem by importing unbiased noisy images and auxiliary features to neural networks to either predict a fixed-sized kernel for convolution or directly predict the denoised result. Since it is impossible to produce arbitrarily high spp images as the training dataset, the network-based denoising fails to produce high-quality images under high spp. More specifically, network-based denoising is inconsistent and does not converge to the ground truth as the sampling rate increases. On the other hand, the post-correction estimators yield a blending coefficient for a pair of biased and unbiased images influenced by image errors or variances to ensure the consistency of the denoised image. As the sampling rate increases, the blending coefficient of the unbiased image converges to 1, that is, using the unbiased image as the denoised results. However, these estimators usually produce artifacts due to the difficulty of accurately predicting image errors or variances with low spp. To address the above problems, we take advantage of both kernel-predicting methods and post-correction denoisers. A novel kernel-based denoiser is proposed based on distribution-free kernel regression consistency theory, which does not explicitly combine the biased and unbiased results but constrain the kernel bandwidth to produce consistent results under high spp. Meanwhile, our kernel regression method explores bandwidth optimization in the robust auxiliary feature space instead of the noisy image space. This leads to consistent high-quality denoising at both low and high spp. Experiment results demonstrate that our method outperforms existing denoisers in accuracy and consistency.

中文翻译：

---

一致 Monte Carlo 降噪的神经核回归


在真实渲染中广泛使用的无偏差 Monte Carlo 路径追踪会产生不需要的杂色，尤其是在每像素采样数 （spp） 较低的情况下。最近，有几种方法解决了这个问题，方法是将无偏噪声图像和辅助特征导入神经网络，以预测固定大小的核进行卷积，或直接预测去噪的结果。由于无法生成任意高 spp 图像作为训练数据集，因此基于网络的去噪无法在高 spp 下产生高质量的图像。更具体地说，基于网络的降噪是不一致的，并且不会随着采样率的增加而收敛到基本实况。另一方面，校正后估计器为一对受图像误差或方差影响的偏置和无偏置图像产生混合系数，以确保去噪图像的一致性。随着采样率的增加，无偏图像的混合系数收敛到 1，即使用无偏图像作为去噪结果。然而，由于难以准确预测图像误差或低 spp 的方差，这些估计器通常会产生伪影。为了解决上述问题，我们利用了核预测方法和校正后降噪器。该文基于无分布核回归一致性理论提出了一种新的基于核的降噪器，该理论没有明确地将有偏和无偏的结果结合起来，而是限制了核带宽，以便在高 spp 下产生一致的结果。同时，我们的核回归方法探索了鲁棒辅助特征空间而不是噪声图像空间中的带宽优化。这导致在低 spp 和高 spp 下都能实现一致的高质量去噪。 实验结果表明，我们的方法在准确性和一致性方面优于现有的降噪器。