ViT Video Game Classification Dataset

10000+ video game frames/images/screenshots across 10 genres (RPG, FPS, MOBA, Horror, Strategy, Roguelike, Platformer, Underwater, Racing, Fighting) for vision transformer (ViT) training.Note: There has been no data augmentation applied for this dataset. To prevent overfitting, it's recommended that you apply respective techniques after extracting the contents of the .zip file. For example, when defining the transforms applied to the images, you could do something like this: transform = transforms.Compose([ transforms.Resize((256, 256)), # ensure images are resized to 256x256 transforms.RandomCrop(224), # random crop to 224x224 transforms.RandomHorizontalFlip(), transforms.RandomRotation(15), transforms.ColorJitter(brightness = 0.2, contrast = 0.2, saturation = 0.2, hue = 0.1), transforms.ToTensor(), transforms.Normalize(mean = [0.485, 0.456, 0.406], std = [0.229, 0.224, 0.225]), transforms.RandomErasing(p = 0.5, scale = (0.02, 0.1)) # for a cutout effect ])