基于特征共生损失的迭代梯度编码网络消除单幅图像反射
Iterative Gradient Encoding Network with Feature Co-Occurrence Loss for Single Image Reflection Removal
摘要:
从在玻璃前拍摄的照片中去除不期望的反射对于提高视觉计算系统的效率非常重要。以前的基于学习的方法已经为某些反射类型产生了视觉上合理的结果,但是,未能对其他反射类型进行概括。有一个关于单图像反射去除,这可以推广到大规模的反射类型的有效方法的文献匮乏。在这项研究中,我们提出了一个迭代梯度编码网络的单一图像反射消除。接下来,为了进一步监督网络在学习传输层特征之间的相关性,我们提出了一种特征同现损失。在SIR $^2$的公共基准数据集上进行的大量实验表明,我们的方法可以在所有成像设置(包括不同背景)上与现有的最先进的方法相比有利地去除反射。此外,随着反射强度的增加,我们的方法仍然可以去除反射,即使其他最先进的方法失败。
英文摘要:
Removing undesired reflections from a photo taken in front of glass is of great importance for enhancing visual computing systems' efficiency. Previous learning-based approaches have produced visually plausible results for some reflections type, however, failed to generalize against other reflection types. There is a dearth of literature for efficient methods concerning single image reflection removal, which can generalize well in large-scale reflection types. In this study, we proposed an iterative gradient encoding network for single image reflection removal. Next, to further supervise the network in learning the correlation between the transmission layer features, we proposed a feature co-occurrence loss. Extensive experiments on the public benchmark dataset of SIR$^2$ demonstrated that our method can remove reflection favorably against the existing state-of-the-art method on all imaging settings, including diverse backgrounds. Moreover, as the reflection strength increases, our method can still remove reflection even where other state of the art methods failed.
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。