Fusing Global and Local Features for Generalized AI-Synthesized Image Detection

2022-03-26 · Yan Ju, Shan Jia, Lipeng Ke, Hongfei Xue, Koki Nagano, Siwei Lyu

With the development of the Generative Adversarial Networks (GANs) and DeepFakes, AI-synthesized images are now of such high quality that humans can hardly distinguish them from real images. It is imperative for media forensics to develop detectors to expose them accurately. Existing detection methods have shown high performance in generated images detection, but they tend to generalize poorly in the real-world scenarios, where the synthetic images are usually generated with unseen models using unknown source data. In this work, we emphasize the importance of combining information from the whole image and informative patches in improving the generalization ability of AI-synthesized image detection. Specifically, we design a two-branch model to combine global spatial information from the whole image and local informative features from multiple patches selected by a novel patch selection module. Multi-head attention mechanism is further utilized to fuse the global and local features. We collect a highly diverse dataset synthesized by 19 models with various objects and resolutions to evaluate our model. Experimental results demonstrate the high accuracy and good generalization ability of our method in detecting generated images. Our code is available at https://github.com/littlejuyan/FusingGlobalandLocal.

📄 PDF Abstract BibTeX arXiv:2203.13964

Code (1)

littlejuyan/fusingglobalandlocal 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.

Federated and Generalized Person Re-identification through Domain and Feature Hallucinating

2022-03-05 · Fengxiang Yang, Zhun Zhong, Zhiming Luo, Shaozi Li 외

In this paper, we study the problem of federated domain generalization (FedDG) for person re-identification (re-ID), which aims to learn a generalized model with multiple decentralized labeled source domains. An empirica…

Domain Generalization Person Re-Identification Unsupervised Domain Adaptation

Alleviating Feature Confusion for Generative Zero-shot Learning

2019-09-17 · Jingjing Li, Mengmeng Jing, Ke Lu, Lei Zhu 외

Lately, generative adversarial networks (GANs) have been successfully applied to zero-shot learning (ZSL) and achieved state-of-the-art performance. By synthesizing virtual unseen visual features, GAN-based methods conve…

Generalized Zero-Shot Learning Zero-Shot Learning

Designing and Training of A Dual CNN for Image Denoising

2020-07-08 · Chunwei Tian, Yong Xu, WangMeng Zuo, Bo Du 외

Deep convolutional neural networks (CNNs) for image denoising have recently attracted increasing research interest. However, plain networks cannot recover fine details for a complex task, such as real noisy images. In th…

Denoising Image Denoising

GLFF: Global and Local Feature Fusion for AI-synthesized Image Detection

2022-11-16 · Yan Ju, Shan Jia, Jialing Cai, Haiying Guan 외

With the rapid development of deep generative models (such as Generative Adversarial Networks and Diffusion models), AI-synthesized images are now of such high quality that humans can hardly distinguish them from pristin…

Federated Unsupervised Visual Representation Learning via Exploiting General Content and Personal Style

2022-11-11 · Yuewei Yang, Jingwei Sun, Ang Li, Hai Li 외

Discriminative unsupervised learning methods such as contrastive learning have demonstrated the ability to learn generalized visual representations on centralized data. It is nonetheless challenging to adapt such methods…

Contrastive Learning Federated Learning Representation Learning