Multi-Supervised Encoder-Decoder for Image Forgery Localization

Image manipulation localization is one of the most challenging tasks because it pays more attention to tampering artifacts than to image content, which suggests that richer features need to be learned. Unlike many existing solutions, we employ a semantic segmentation network, named Multi-Supervised...

Full description

Saved in:
Bibliographic Details
Published inElectronics (Basel) Vol. 10; no. 18; p. 2255
Main Authors Yu, Chunfang, Zhou, Jizhe, Li, Qin
Format Journal Article
LanguageEnglish
Published Basel MDPI AG 01.09.2021
Subjects
Ablation
Classification
Coders
Encoders-Decoders
Forgery
Image manipulation
Image segmentation
Localization
Neural networks
Noise
Semantics
Spatial data
Training
Online AccessGet full text

Cover

More Information
Summary:Image manipulation localization is one of the most challenging tasks because it pays more attention to tampering artifacts than to image content, which suggests that richer features need to be learned. Unlike many existing solutions, we employ a semantic segmentation network, named Multi-Supervised Encoder–Decoder (MSED), for the detection and localization of forgery images with arbitrary sizes and multiple types of manipulations without extra pre-training. In the basic encoder–decoder framework, the former encodes multi-scale contextual information by atrous convolution at multiple rates, while the latter captures sharper object boundaries by applying upsampling to gradually recover the spatial information. The additional multi-supervised module is designed to guide the training process by multiply adopting pixel-wise Binary Cross-Entropy (BCE) loss after the encoder and each upsampling. Experiments on four standard image manipulation datasets demonstrate that our MSED network achieves state-of-the-art performance compared to alternative baselines.
ISSN:2079-9292
2079-9292
DOI:10.3390/electronics10182255