191578010 / dsgcg Goto Github PK

Download link：https://pan.baidu.com/s/1Ke1A6UIRYGn0gYJd5SvtBQ?pwd=gcgd 
password：gcgd 

There are 42000 CG images and 42000 PG images sized from 256 × 256 to 4928 × 3264 with moderate
to good visual quality in the DSGCG dataset. The CG set contains traditional CG images, GAN images, and DM images.

6000 TCG image set were constructed by collecting CG video or game screenshots (namely Forza Horizon, GANTZ,

God of War, Red Dead Redemption, and Playerunknown’s Battlegrounds) from Kaggle and Google websites. We used

the pretrained StyleGAN2, ProGAN, and BigGAN models for GAN images generation (Total number: 18000,

each GAN contains 6000 images with 6 categories). Stable diffusion, Latent diffusion, and DALL · E Mini

were adopted to generate 18000 (=6000+6000+6000) DM images, respectively. For images in the PG set, 402 images

with different contents were taken by the authors using a NIKON D5200 camera, the focal length and exposure time

are 25mm and 1/200 second respectively. 1600 and 5998 images were downloaded from the Columbia and RAISE

datasets. 24000 were downloaded from LSUN dataset. Since grayscale images are also widely used in real
scenario, 10000 grayscale PG images in BOSSbase v1.01 dataset were also collected to enhance the diversity of

images.

Supplementary data set： contains 40000 DM images generated by several state-of-the-art pre-trained generative models, including Stable Diffusion \cite{Stable}, DALL-E Mini \cite{Dayma2021dall}, Latent Diffusion \cite{rombach2022high}, and GLIDE \cite{nichol2021glide} models.