arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2301.13554cs.CV
收藏

NoiseTransfer:使用对比嵌入生成图像噪声

NoiseTransfer: Image Noise Generation with Contrastive Embeddings

Hanyang University(汉阳大学)

Seunghwan Lee, Tae Hyun Kim

摘要:

在大量合成训练数据集的帮助下,深度图像去噪网络取得了令人印象深刻的成功。然而,真实世界的去噪仍然是一个具有挑战性的问题,由于真实和合成噪声数据集之间的分布的差异。尽管已经提出了几个真实世界的噪声数据集,但是训练数据集的数量(即,干净和真实噪声图像对)是有限的,并且获取更多真实噪声数据集是费力且昂贵的。为了缓解这个问题,已经研究了使用生成模型来模拟真实噪声模型的许多尝试。然而,以前的工作必须训练多个网络来处理多个不同的噪声分布。相比之下,我们提出了一个新的生成模型,可以合成多个不同的噪声分布的噪声图像。具体来说,我们采用最近的对比学习来学习噪声的可区分的潜在特征。此外,我们的模型可以产生新的噪声图像,通过转移的噪声特性仅从一个单一的参考噪声图像。我们证明了我们的噪声模型的准确性和有效性,已知和未知的噪声去除。

英文摘要:

Deep image denoising networks have achieved impressive success with the help of a considerably large number of synthetic train datasets. However, real-world denoising is a still challenging problem due to the dissimilarity between distributions of real and synthetic noisy datasets. Although several real-world noisy datasets have been presented, the number of train datasets (i.e., pairs of clean and real noisy images) is limited, and acquiring more real noise datasets is laborious and expensive. To mitigate this problem, numerous attempts to simulate real noise models using generative models have been studied. Nevertheless, previous works had to train multiple networks to handle multiple different noise distributions. By contrast, we propose a new generative model that can synthesize noisy images with multiple different noise distributions. Specifically, we adopt recent contrastive learning to learn distinguishable latent features of the noise. Moreover, our model can generate new noisy images by transferring the noise characteristics solely from a single reference noisy image. We demonstrate the accuracy and the effectiveness of our noise model for both known and unknown noise removal.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑