CoolFace
Datasetpublic

TheKernel01/AIGC-Detection-Benchmark

AIGC Detection Benchmark Dataset 📝 Dataset Description Dataset Summary The AIGC Detection Benchmark Dataset is a high-quality collection of images and associated metadata designed to benchmark models for detecting and identifying the source of artificially generated content. The dataset contains a mix of real-world images and images generated by a wide array of prominent AI models, including diffusion models (like Stable Diffusion, DALL-E 2, Midjourney, ADM) and… See the full description on the dataset page: https://huggingface.co/datasets/TheKernel01/AIGC-Detection-Benchmark.

sourceHugging Faceapache-2.0updated 6mo agoView on Hugging Face
0likes1.1kdownloads
Dataset Card

AIGC Detection Benchmark Dataset

📝 Dataset Description

Dataset Summary

The AIGC Detection Benchmark Dataset is a high-quality collection of images and associated metadata designed to benchmark models for detecting and identifying the source of artificially generated content. The dataset contains a mix of real-world images and images generated by a wide array of prominent AI models, including diffusion models (like Stable Diffusion, DALL-E 2, Midjourney, ADM) and GANs (like BigGAN, StyleGAN, ProGAN).

Each image is meticulously labeled under two categories, enabling researchers to tackle two distinct, high-value computer vision tasks: binary real/fake classification and multi-class source model identification. Note: This specific version of the dataset is designed exclusively for testing and evaluation purposes, with all data consolidated into a single test split.

Supported Tasks and Leaderboards

This dataset directly supports two critical image classification tasks:

**Task ID****Task Name****Description****Output Classes**
Task ABinary Veracity ClassificationClassifying images as either real or fake.2 (real, fake)
Task BAI Model Source IdentificationIdentifying the specific AI generation model used for images labeled as AI-Generated.18 (Real, ADM, BigGAN, CycleGAN, DALLE2, GauGAN, GLIDE, Midjourney, ProGAN, SD14, SD15, SDXL, StarGAN, StyleGAN, StyleGAN2, VQDM, WhichFaceIsReal, Wukong)

Languages

The descriptive text, including all captions, is in English (en).

🗂️ Data Splits

All instances have been merged into a single test split to serve strictly as an evaluation benchmark.

**Split****Number of Instances****Notes**
test125,026Used exclusively for final, unbiased model evaluation and benchmarking.

💾 Dataset Structure

Data Instances

A single data instance consists of an image file and two distinct labels detailing its source and authenticity.

**Field Name****Example Value****Description**
image<PIL.Image.Image object>The actual image content loaded into a PIL object.
label1Binary label for authenticity (Real vs. AI-Generated).
generator3Multi-class label for the specific generation model (or Real).

Data Fields

The dataset contains the following fields:

**Field Name****Data Type****Description**
imagedatasets.Image()The actual image content (e.g., .jpg, .png).
labeldatasets.ClassLabelTask A: Binary label for image veracity.
generatordatasets.ClassLabelTask B: Label specifying the generation source/model.

🏷️ Label Definitions

The two label fields use the following mappings:

`label` (Binary Veracity Classification)

**Label****Value****Description**
real0Image is a real photograph/non-AI generated.
fake1Image was created by an AI generation model.

`generator` (Model Source Identification)

**Label****Value****Description**
Real0Real image (no AI generation involved).
ADM1Generated by Ablated Diffusion Model (Guided Diffusion).
BigGAN2Generated by BigGAN.
CycleGAN3Generated by CycleGAN.
DALLE24Generated by OpenAI's DALL-E 2.
GauGAN5Generated by GauGAN (SPADE).
GLIDE6Generated by GLIDE.
Midjourney7Generated by Midjourney.
ProGAN8Generated by ProGAN (Progressive GAN).
SD149Generated by Stable Diffusion 1.4.
SD1510Generated by Stable Diffusion 1.5.
SDXL11Generated by Stable Diffusion XL.
StarGAN12Generated by StarGAN.
StyleGAN13Generated by StyleGAN.
StyleGAN214Generated by StyleGAN2.
VQDM15Generated by Vector Quantized Diffusion Model.
WhichFaceIsReal16Real human face sourced from the WhichFaceIsReal dataset.
Wukong17Generated by the Wukong diffusion model.

🔗 Sources