Deep Image Comparator: Learning to Visualize Editorial Change
Deep Image Comparator: Learning to Visualize Editorial Change
复制标题
DOI：
10.1109/cvprw53098.2021.00108
复制
发表时间：
2021-06
期刊：
影响因子：
--
通讯作者：
Alexander Black;Tu Bui;Hailin Jin;Vishy Swaminathan;J. Collomosse
中科院分区：
文献类型：
--
作者：
Alexander Black;Tu Bui;Hailin Jin;Vishy Swaminathan;J. Collomosse
我们提出了一种新的体系结构，用于比较一对图像以识别已经受到编辑操作的图像区域。我们首先描述了一种稳健的近重复搜索，用于将在线传播的潜在篡改图像与可信原件数据库中的图像进行匹配。然后，我们描述了一种新的体系结构，用于比较该图像对，以定位已被操纵以与检索到的原始图像不同的区域。本地化忽略了由于在线重新分发期间经常发生的良性图像转换而导致的差异。这些伪像包括由于噪声和重新压缩降级而产生的伪像，以及由于图像填充、扭曲以及大小和形状的改变而产生的错位变换。通过比较器体系结构内的可微翘曲模块的端到端训练来实现对错位变换的稳健性。在数百万张照片的数据集上，我们展示了对良性变换和处理的图像的有效检索和比较。
We present a novel architecture for comparing a pair of images to identify image regions that have been subjected to editorial manipulation. We first describe a robust near-duplicate search, for matching a potentially manipulated image circulating online to an image within a trusted database of originals. We then describe a novel architecture for comparing that image pair, to localize regions that have been manipulated to differ from the retrieved original. The localization ignores discrepancies due to benign image transformations that commonly occur during online redistribution. These include artifacts due to noise and recompression degradation, as well as out-of-place transformations due to image padding, warping, and changes in size and shape. Robustness towards out-of-place transformations is achieved via the end-to-end training of a differentiable warping module within the comparator architecture. We demonstrate effective retrieval and comparison of benign transformed and manipulated images, over a dataset of millions of photographs.