In machine learning research, using synthetic images (like those from a video game) to train neural networks can be more efficient than using real-world images. That’s because synthetic image data is already labeled and annotated, while real-world image data requires somebody to exhaustively label everything the computer is seeing — that’s a tree, a dog, a bike. But the synthetic image approach can be problematic as what the algorithm learns doesn’t always carry over neatly to real world scenes. The synthetic image data “is often not realistic enough, leading the network to learn details only present in synthetic images and fail to generalize well on real images,” the paper from Apple says.