{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/introvae-introspective-variational","title":"IntroVAE: Introspective Variational Autoencoders for Photographic Image Synthesis","arxiv_id":"1807.06358","date":"2018-07-17","proceeding":"NeurIPS 2018 12","authors":["Huaibo Huang","Zhihang Li","Ran He","Zhenan Sun","Tieniu Tan"],"abstract":"We present a novel introspective variational autoencoder (IntroVAE) model for\nsynthesizing high-resolution photographic images. IntroVAE is capable of\nself-evaluating the quality of its generated samples and improving itself\naccordingly. Its inference and generator models are jointly trained in an\nintrospective way. On one hand, the generator is required to reconstruct the\ninput images from the noisy outputs of the inference model as normal VAEs. On\nthe other hand, the inference model is encouraged to classify between the\ngenerated and real samples while the generator tries to fool it as GANs. These\ntwo famous generative frameworks are integrated in a simple yet efficient\nsingle-stream architecture that can be trained in a single stage. IntroVAE\npreserves the advantages of VAEs, such as stable training and nice latent\nmanifold. Unlike most other hybrid models of VAEs and GANs, IntroVAE requires\nno extra discriminators, because the inference model itself serves as a\ndiscriminator to distinguish between the generated and real samples.\nExperiments demonstrate that our method produces high-resolution\nphoto-realistic images (e.g., CELEBA images at \\(1024^{2}\\)), which are\ncomparable to or better than the state-of-the-art GANs.","url_abs":"http://arxiv.org/abs/1807.06358v2","url_pdf":"http://arxiv.org/pdf/1807.06358v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"introvae-introspective-variational","repo_url":"https://github.com/barisbatuhan/SSuperGAN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"introvae-introspective-variational","repo_url":"https://github.com/bbeatrix/introvae","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"introvae-introspective-variational","repo_url":"https://github.com/bogedy/intro_dfc","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"}],"methods":[{"method_slug":"affine-coupling","method_name":"Affine Coupling"},{"method_slug":"normalizing-flows","method_name":"Normalizing Flows"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1807.06358","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}