ObjectFormer for Image Manipulation Detection and Localization
This addresses the challenge of ensuring trustworthiness in multimedia data for applications like forensics and security, representing a strong incremental improvement.
The paper tackles the problem of detecting and localizing image manipulations by proposing ObjectFormer, which extracts high-frequency features and uses object prototypes to model consistencies, achieving state-of-the-art results on various datasets.
Recent advances in image editing techniques have posed serious challenges to the trustworthiness of multimedia data, which drives the research of image tampering detection. In this paper, we propose ObjectFormer to detect and localize image manipulations. To capture subtle manipulation traces that are no longer visible in the RGB domain, we extract high-frequency features of the images and combine them with RGB features as multimodal patch embeddings. Additionally, we use a set of learnable object prototypes as mid-level representations to model the object-level consistencies among different regions, which are further used to refine patch embeddings to capture the patch-level consistencies. We conduct extensive experiments on various datasets and the results verify the effectiveness of the proposed method, outperforming state-of-the-art tampering detection and localization methods.