[论文解读] Blind Deblurring Using GANs

Manoj Kumar Lenka, Anubha Pandey+1 more|arXiv (Cornell University)|Jul 27, 2019

Advanced Image Processing Techniques参考文献 17被引用 5

一句话总结

本文提出一种基于GAN的盲图像去模糊方法，通过注意力模块、残差连接和多损失训练增强全局感知能力。该方法在GoPro数据集上提升了去模糊性能，DeblurGAN模型的PSNR达到28.70，SSIM达到0.958，进一步结合自注意力机制与感知损失后性能进一步提升。

ABSTRACT

Deblurring is the task of restoring a blurred image to a sharp one, retrieving the information lost due to the blur. In blind deblurring we have no information regarding the blur kernel. As deblurring can be considered as an image to image translation task, deep learning based solutions, including the ones which use GAN (Generative Adversarial Network), have been proven effective for deblurring. Most of them have an encoder-decoder structure. Our objective is to try different GAN structures and improve its performance through various modifications to the existing structure for supervised deblurring. In supervised deblurring we have pairs of blurred and their corresponding sharp images, while in the unsupervised case we have a set of blurred and sharp images but their is no correspondence between them. Modifications to the structures is done to improve the global perception of the model. As blur is non-uniform in nature, for deblurring we require global information of the entire image, whereas convolution used in CNN is able to provide only local perception. Deep models can be used to improve global perception but due to large number of parameters it becomes difficult for it to converge and inference time increases, to solve this we propose the use of attention module (non-local block) which was previously used in language translation and other image to image translation tasks in deblurring. Use of residual connection also improves the performance of deblurring as features from the lower layers are added to the upper layers of the model. It has been found that classical losses like L1, L2, and perceptual loss also help in training of GANs when added together with adversarial loss. We also concatenate edge information of the image to observe its effects on deblurring. We also use feedback modules to retain long term dependencies

研究动机与目标

• 提升基于GAN的去模糊模型在非均匀模糊场景下的全局感知能力，以实现对整幅图像的上下文理解。
• 通过引入注意力机制，弥补CNN局部感受野的局限性，以建模长距离依赖关系。
• 通过结合对抗损失与L1、L2及感知损失，提升训练稳定性和模型性能。
• 评估残差连接、边缘信息与反馈模块在提升去模糊质量方面的有效性。
• 通过架构改进与损失函数优化，提升现有GAN架构（Pix2Pix、DeblurGAN、RiR）在清晰度与结构相似性方面的表现。

提出的方法

• 在每个编码器/解码器块之后每隔三个块集成非局部自注意力模块，以捕捉长距离依赖关系并改善全局上下文建模。
• 应用通道注意力机制，对特征图进行重校准，聚焦于显著通道以提升表征学习能力。
• 采用全局残差连接，保留深层网络中的低级特征，促进梯度流动并提升特征复用。
• 结合对抗损失与基于VGG特征的感知损失、L1和L2损失，以稳定训练过程并提升视觉质量。
• 在判别器中引入谱归一化，以稳定GAN训练并防止模式崩溃。
• 引入反馈模块，通过多轮迭代逐步优化去模糊输出，以保留长期依赖关系。

实验结果

研究问题

• RQ1 注意力机制是否能显著提升GAN在盲图像去模糊任务中的全局感知能力？
• RQ2 残差连接与特征复用在深层编码器-解码器架构中对去模糊性能有何影响？
• RQ3 结合多种损失函数（对抗损失、感知损失、L1/L2损失）是否能带来更好的收敛性与更清晰的输出？
• RQ4 边缘信息与反馈模块对基于GAN的去模糊模型质量有何影响？
• RQ5 架构选择（如更深网络、跳跃连接）如何影响盲去模糊任务中的PSNR与SSIM？

主要发现

• DeblurGAN模型在GoPro数据集上达到PSNR 28.70与SSIM 0.958，优于基线模型。
• 在Pix2Pix中引入自注意力与感知损失后，PSNR从25.41提升至27.29，SSIM从0.810提升至0.858。
• 带有注意力与感知损失的Residual-in-Residual（RiR）模型在高分辨率1280×768图像上达到PSNR 23.46。
• 反馈模块导致性能下降，PSNR降至27.20，SSIM降至0.827，表明在此设置下收益有限。
• 边缘信息的引入降低了性能，PSNR降至25.27，SSIM降至0.773，表明其在本任务中可能无效。
• 谱归一化与多损失训练显著提升了所有架构的训练稳定性与最终模型性能。