arXiv (Cornell University) · 2018 · 164 citations · 0 references
ComputingComputer ScienceComputer VisionMachine LearningRenderingDeep LearningGenerative ArchitectureGenerative ModelingImage SynthesisVariational InferenceAutoencodersGenerative Adversarial NetworksGenerative Image SynthesisPhoto-realistic RenderingVideo Generation
IntroVAE retains the stable training and smooth latent manifold benefits of traditional VAEs. The paper introduces IntroVAE, a novel introspective VAE designed to synthesize high‑resolution photographic images. IntroVAE jointly trains its inference and generator in a single‑stream architecture that blends VAE reconstruction with GAN‑style adversarial training, using the inference model as a discriminator. Experiments on CELEBA at 1024×1024 show that IntroVAE produces photo‑realistic images that match or surpass the quality of leading GANs.
We present a novel introspective variational autoencoder (IntroVAE) model for synthesizing high-resolution photographic images. IntroVAE is capable of self-evaluating the quality of its generated samples and improving itself accordingly. Its inference and generator models are jointly trained in an introspective way. On one hand, the generator is required to reconstruct the input images from the noisy outputs of the inference model as normal VAEs. On the other hand, the inference model is encouraged to classify between the generated and real samples while the generator tries to fool it as GANs. These two famous generative frameworks are integrated in a simple yet efficient single-stream architecture that can be trained in a single stage. IntroVAE preserves the advantages of VAEs, such as stable training and nice latent manifold. Unlike most other hybrid models of VAEs and GANs, IntroVAE requires no extra discriminators, because the inference model itself serves as a discriminator to distinguish between the generated and real samples. Experiments demonstrate that our method produces high-resolution photo-realistic images (e.g., CELEBA images at \(1024^{2}\)), which are comparable to or better than the state-of-the-art GANs.