Chengpong/aigcdetect_test

该数据集是一个包含多个生成模型生成的图像集合，用于区分真实与生成图像的任务。数据集包括多个分割，每个分割对应不同的生成模型或来源，如ADM、DALLE2、Glide、Midjourney、VQDM、biggan、cyclegan、gaugan、progan、sd_xl、stable_diffusion_v_1_4、stable_diffusion_v_1_5、stargan、stylegan、stylegan2、whichfaceisreal和wukong。每个图像都有RGB模式和是否真实的布尔标签（is_real）。数据集总大小约为36.55 GB，包含超过150,000个示例，覆盖了多种图像生成技术。This dataset is a collection of images generated by multiple generative models, designed for tasks distinguishing real from generated images. It includes multiple splits, each corresponding to different generative models or sources, such as ADM, DALLE2, Glide, Midjourney, VQDM, biggan, cyclegan, gaugan, progan, sd_xl, stable_diffusion_v_1_4, stable_diffusion_v_1_5, stargan, stylegan, stylegan2, whichfaceisreal, and wukong. Each image has RGB mode and a boolean label indicating whether it is real (is_real). The total dataset size is approximately 36.55 GB, with over 150,000 examples, covering a variety of image generation techniques.