Unconditional Image Generation
21 papers with code • 3 benchmarks • 3 datasets
By decomposing the image formation process into a sequential application of denoising autoencoders, diffusion models (DMs) achieve state-of-the-art synthesis results on image data and beyond.
Combined with multiple architectural improvements, we achieve record-breaking performance for unconditional image generation on CIFAR-10 with an Inception score of 9. 89 and FID of 2. 20, a competitive likelihood of 2. 99 bits/dim, and demonstrate high fidelity generation of 1024 x 1024 images for the first time from a score-based generative model.
We extensively evaluate the representation learning and generation capabilities of these BigBiGAN models, demonstrating that these generation-based models achieve the state of the art in unsupervised representation learning on ImageNet, as well as in unconditional image generation.
Neural architecture search (NAS) has witnessed prevailing success in image classification and (very recently) segmentation tasks.
We propose a manifold matching approach to generative models which includes a distribution generator (or data generator) and a metric generator.
In this paper, we propose an AdversarialNAS method specially tailored for Generative Adversarial Networks (GANs) to search for a superior generative model on the task of unconditional image generation.
While Generative Adversarial Networks (GANs) show increasing performance and the level of realism is becoming indistinguishable from natural images, this also comes with high demands on data and computation.
Such a model disentangles latent factors according to the semantic of feature channels by channel-/group- wise fusion of latent codes and feature channels.