ImagiNet：通過對比學習實現通用合成圖像檢測的多內容數據集

ImagiNet: A Multi-Content Dataset for Generalizable Synthetic Image Detection via Contrastive Learning

July 29, 2024

作者: Delyan Boychev, Radostin Cholakov

cs.AI

摘要

生成模型，如擴散模型（DMs）、變分自編碼器（VAEs）和生成對抗網絡（GANs），能夠生成具有幾乎與真實照片和藝術品難以區分的真實性水平的圖像。儘管這種能力對許多行業都有益，但識別合成圖像的困難使在線媒體平台容易受到冒充和虛假信息的威脅。為了支持防禦方法的發展，我們引入了ImagiNet，這是一個用於合成圖像檢測的高分辨率和平衡數據集，旨在減輕現有資源中的潛在偏見。它包含20萬個示例，涵蓋四個內容類別：照片、繪畫、人臉和未分類。合成圖像是使用開源和專有生成器生成的，而相同內容類型的真實對應則是從公共數據集中收集的。ImagiNet的結構允許建立一個雙軌評估系統：i）分類為真實或合成，ii）識別生成模型。為了建立基準，我們使用自監督對比目標（SelfCon）訓練了一個ResNet-50模型，分別用於每個軌道。該模型在已建立的基準測試中展示出最先進的性能和高推理速度，實現了高達0.99的AUC和在86%至95%之間的平衡準確度，即使在涉及壓縮和調整大小的社交網絡條件下也是如此。我們的數據和代碼可在https://github.com/delyan-boychev/imaginet 上獲得。

English

Generative models, such as diffusion models (DMs), variational autoencoders (VAEs), and generative adversarial networks (GANs), produce images with a level of authenticity that makes them nearly indistinguishable from real photos and artwork. While this capability is beneficial for many industries, the difficulty of identifying synthetic images leaves online media platforms vulnerable to impersonation and misinformation attempts. To support the development of defensive methods, we introduce ImagiNet, a high-resolution and balanced dataset for synthetic image detection, designed to mitigate potential biases in existing resources. It contains 200K examples, spanning four content categories: photos, paintings, faces, and uncategorized. Synthetic images are produced with open-source and proprietary generators, whereas real counterparts of the same content type are collected from public datasets. The structure of ImagiNet allows for a two-track evaluation system: i) classification as real or synthetic and ii) identification of the generative model. To establish a baseline, we train a ResNet-50 model using a self-supervised contrastive objective (SelfCon) for each track. The model demonstrates state-of-the-art performance and high inference speed across established benchmarks, achieving an AUC of up to 0.99 and balanced accuracy ranging from 86% to 95%, even under social network conditions that involve compression and resizing. Our data and code are available at https://github.com/delyan-boychev/imaginet.