CoChat
Sign Up
Cited by 2,514Generative Adversarial Networks and Image SynthesisAdvanced Neural Network ApplicationsMultimodal Machine Learning Applications

Taming Transformers for High-Resolution Image Synthesis

Open access
Esser, Patrick, Rombach, Robin, Ommer, Bjorn·2021·2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
DOI 10.1109/cvpr46437.2021.01268
Open this paper in CoChat — free accountOpen PDF

Abstract

Designed to learn long-range interactions on sequential data, transformers continue to show state-of-the-art results on a wide variety of tasks. In contrast to CNNs, they contain no inductive bias that prioritizes local interactions. This makes them expressive, but also computationally infeasible for long sequences, such as high-resolution images. We demonstrate how combining the effectiveness of the inductive bias of CNNs with the expressivity of transformers enables them to model and thereby synthesize high-resolution images. We show how to (i) use CNNs to learn a context-rich vocabulary of image constituents, and in turn (ii) utilize transformers to efficiently model their composition within high-resolution images. Our approach is readily applied to conditional synthesis tasks, where both non-spatial information, such as object classes, and spatial information, such as segmentations, can control the generated image. In particular, we present the first results on semantically-guided synthesis of megapixel images with transformers. Project page at https://git.io/JLlvY.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/cvpr46437.2021.01268

Cites

8 of 8 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Transformer language models and other latest generative approaches such as VQ-VAE, VQGAN + autoregressive transformer, etc., will definitely supplement and enhance DeepFake generation techniques.

    “They,together with other latest generative approaches such as vector quantized variational autoencoder (VQ-VAE) (Razavi et al. 2019), VQGAN + autoregressive transformer (Esser et al. 2021), etc., will definitely supplement and enhance the DeepFake generation techniques.”

    From Countering Malicious DeepFakes: Survey, Battleground, and Horizon (2022) · cites this paper as Esser et al. 2021

  • Unified models for understanding and generation such as Chameleon employ the discrete image tokenizer VQGAN for image interpretation.

    “Unified models for understanding and generation, such as Chameleon (Team, 2024), employ the discrete image tokenizer VQGAN (Esser et al., 2021) for image interpretation.”

    From WorldVLA: Towards Autoregressive Action World Model (2025) · cites this paper as Esser et al., 2021

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 52

  1. CoRR, abs/2010.02917, 2020. 8 · arXiv
  2. [3]Yoshua Bengio, Nicholas Léonard, and Aaron C. Courville. Estimating or propagating gradients through stochastic neurons for conditional computation. CoRR, abs/1308.3432, 2013. 4 · arXiv
  3. International Conference on Learning Representations, ICLR, 2019. 8, 10, 16, 17, 18, 19
  4. Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel M. Ziegler, Jeffrey Wu, Clemens Winter, Christopher Hesse, Mark Chen, Eric Sigler, Mateusz Litwin, Scott Gray, Benjamin Chess, Jack Clark, Christopher Berner, Sam McCandlish, Alec Radford, Ilya Sutskever, and Dario Amodei. Language Models are Few-Shot Learners. arXiv preprint arXiv:2005.14165, 2020. 1 · arXiv
  5. [6]Holger Caesar, Jasper Uijlings, and Vittorio Ferrari. COCO-Stuff: Thing and stuff classes in context. In Computer vision and pattern recognition (CVPR), 2018 IEEE conference on. IEEE, 2018. 6
  6. [7]Liang-Chieh Chen, G. Papandreou, I. Kokkinos, Kevin Murphy, and A. Yuille. DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018. 5
  7. [10]Xi Chen, Nikhil Mishra, Mostafa Rohaninejad, and Pieter Abbeel. Pixelsnail: An improved autoregressive generative model. In ICML, volume 80 of Proceedings of Machine Learning Research, pages 863–871. PMLR, 2018. 2, 5
  8. [11]Rewon Child. Very deep vaes generalize autoregressive models and can outperform them on images. CoRR, abs/2011.10650, 2020. 8 · arXiv
  9. IEEE Computer Society Conference on Computer Vision and Pattern Recognition CVPR, 2009. 5
  10. [17]Alexey Dosovitskiy and Thomas Brox. Generating Images with Perceptual Similarity Metrics based on Deep Networks. In Advances in Neural Information Processing Systems 29: Annual Conference on Neural Information Processing Systems, NeurIPS, 2016. 4
  11. [18]Patrick Esser, Robin Rombach, and Björn Ommer. A Disentangling Invertible Interpretation Network for Explaining Latent Repre- sentations. In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR, 2020. 2
  12. [19]Jeffrey De Fauw, Sander Dieleman, and Karen Simonyan. Hierarchical autoregressive image models with auxiliary decoders. CoRR, abs/1903.04933, 2019. 16, 17, 18, 19 · arXiv
  13. [20]Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron C. Courville, and Yoshua Bengio. Generative Adversarial Nets. In Advances in Neural Information Processing Systems 27: Annual Conference on Neural Information Processing Systems, NeurIPS, 2014. 2
  14. [21]Seungwook Han, Akash Srivastava, Cole L. Hurwitz, Prasanna Sattigeri, and David D. Cox. not-so-biggan: Generating high-fidelity images on a small compute budget. CoRR, abs/2009.04433, 2020. 2 · arXiv
  15. [23]Ari Heljakka, Arno Solin, and Juho Kannala. Pioneer networks: Progressively growing generative autoencoder. In C. V. Jawahar, Hongdong Li, Greg Mori, and Konrad Schindler, editors, Computer Vision - ACCV 2018 - 14th Asian Conference on Computer Vision, Perth, Australia, December 2-6, 2018, Revised Selected Papers, Part I, 2018. 8
  16. [24]Ari Heljakka, Arno Solin, and Juho Kannala. Towards photographic image manipulation with balanced growing of generative autoencoders. In IEEE Winter Conference on Applications of Computer Vision, WACV 2020, Snowmass Village, CO, USA, March 1-5, 2020, pages 3109–3118. IEEE, 2020. 8
  17. [27]Ari Holtzman, Jan Buys, Li Du, Maxwell Forbes, and Yejin Choi. The curious case of neural text degeneration. In ICLR. OpenRe- view.net, 2020. 8, 22
  18. In 2017 IEEE Conference on Computer Vision and Pattern Recognition, CVPR, 2017. 4, 11
  19. [29]Eric Jang, Shixiang Gu, and Ben Poole. Categorical reparameterization with gumbel-softmax. arXiv preprint arXiv:1611.01144, · arXiv
  20. [30]Justin Johnson, Alexandre Alahi, and Li Fei-Fei. Perceptual losses for real-time style transfer and super-resolution. In ECCV (2), volume 9906 of Lecture Notes in Computer Science, pages 694–711. Springer, 2016. 4
  21. CoRR, abs/1710.10196, 2017. 6, 8 · arXiv
  22. [32]Tero Karras, Miika Aittala, Janne Hellsten, Samuli Laine, Jaakko Lehtinen, and Timo Aila. Training generative adversarial networks with limited data. In Hugo Larochelle, Marc'Aurelio Ranzato, Raia Hadsell, Maria-Florina Balcan, and Hsuan-Tien Lin, editors, Ad- vances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual, 2020. 8
  23. [34]Tero Karras, Samuli Laine, Miika Aittala, Janne Hellsten, Jaakko Lehtinen, and Timo Aila. Analyzing and improving the image quality of stylegan. In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2020, Seattle, WA, USA, June 13-19, 2020, pages 8107–8116. IEEE, 2020. 8
  24. Information Processing Systems 31: Annual Conference on Neural Information Processing Systems 2018, NeurIPS, 2018. 8
  25. [38]Diederik P. Kingma and Max Welling. Auto-Encoding Variational Bayes. In 2nd International Conference on Learning Representa- tions, ICLR, 2014. 2
  26. 6706–6713. AAAI Press, 2019. 2
  27. [42]Chieh Hubert Lin, Chia-Che Chang, Yu-Sheng Chen, Da-Cheng Juan, Wei Wei, and Hwann-Tzong Chen. COCO-GAN: generation by parts via conditional coordinating. In ICCV, pages 4511–4520. IEEE, 2019. 5
  28. [44]Peter J. Liu, Mohammad Saleh, Etienne Pot, Ben Goodrich, Ryan Sepassi, Lukasz Kaiser, and Noam Shazeer. Generating wikipedia by summarizing long sequences. In ICLR (Poster). OpenReview.net, 2018. 4
  29. [45]Ziwei Liu, Ping Luo, Shi Qiu, Xiaogang Wang, and Xiaoou Tang. Deepfashion: Powering robust clothes recognition and retrieval with rich annotations. In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2016. 6
  30. In 7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019. OpenReview.net, 2019. 14
  31. [51]B. Ommer and J. M. Buhmann. Learning the compositional nature of visual objects. In 2007 IEEE Conference on Computer Vision and Pattern Recognition, pages 1–8, 2007. 2
  32. [53]Taesung Park, Ming-Yu Liu, Ting-Chun Wang, and Jun-Yan Zhu. Semantic Image Synthesis with Spatially-Adaptive Normalization. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR, 2019. 2, 7, 42, 43
  33. In ICML, volume 80 of Proceedings of Machine Learning Research, pages 4052–4061. PMLR, 2018. 2, 3, 5
  34. [56]Stanislav Pidhorskyi, Donald A. Adjeroh, and Gianfranco Doretto. Adversarial latent autoencoders. In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2020, Seattle, WA, USA, June 13-19, 2020, pages 14092–14101. IEEE, 2020. 8
  35. Mixing datasets for zero-shot cross-dataset transfer. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2020. 5
  36. [62]Danilo Jimenez Rezende, Shakir Mohamed, and Daan Wierstra. Stochastic backpropagation and approximate inference in deep generative models. In Proceedings of the 31st International Conference on International Conference on Machine Learning, ICML, 2014. 2
  37. [63]Robin Rombach, Patrick Esser, and Björn Ommer. Making sense of cnns: Interpreting deep representations and their invariances with inns. In Andrea Vedaldi, Horst Bischof, Thomas Brox, and Jan-Michael Frahm, editors, Computer Vision - ECCV 2020 - 16th European Conference, Glasgow, UK, August 23-28, 2020, Proceedings, Part XVII, volume 12362 of Lecture Notes in Computer Science, pages 647–664. Springer, 2020. 2
  38. [64]Robin Rombach, Patrick Esser, and Bjorn Ommer. Network-to-network translation with conditional invertible neural networks. In H. Larochelle, M. Ranzato, R. Hadsell, M. F. Balcan, and H. Lin, editors, Advances in Neural Information Processing Systems, volume 33, pages 2784–2797. Curran Associates, Inc., 2020. 2
  39. [65]Shibani Santurkar, Dimitris Tsipras, Brandon Tran, Andrew Ilyas, Logan Engstrom, and Aleksander Madry. Computer vision with a single (robust) classifier. In ArXiv preprint arXiv:1906.09453, 2019. 5 · arXiv
  40. [66]Edgar Schönfeld, Bernt Schiele, and Anna Khoreva. A u-net based discriminator for generative adversarial networks. In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2020, Seattle, WA, USA, June 13-19, 2020, pages 8204– 8213. IEEE, 2020. 8
  41. [68]Aliaksandr Siarohin, Stéphane Lathuilière, Sergey Tulyakov, Elisa Ricci, and Nicu Sebe. First order motion model for image anima- tion. In Conference on Neural Information Processing Systems (NeurIPS), December 2019. 2
  42. Raia Hadsell, Maria-Florina Balcan, and Hsuan-Tien Lin, editors, Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual, 2020. 8
  43. [70]Aäron van den Oord, Nal Kalchbrenner, and Koray Kavukcuoglu. Pixel recurrent neural networks. In ICML, volume 48 of JMLR Workshop and Conference Proceedings, pages 1747–1756. JMLR.org, 2016. 2
  44. [74]Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is All you Need. In Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems, NeurIPS, 2017. 1, 2, 3
  45. [75]Ting-Chun Wang, Ming-Yu Liu, Jun-Yan Zhu, Andrew Tao, Jan Kautz, and Bryan Catanzaro. High-resolution image synthesis and semantic manipulation with conditional gans. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018. 7
  46. [76]Dirk Weissenborn, Oscar Täckström, and Jakob Uszkoreit. Scaling autoregressive video models. In ICLR. OpenReview.net, 2020. 2
  47. [78]Zhisheng Xiao, Qing Yan, Yi-an Chen, and Yali Amit. Generative latent flow: A framework for non-adversarial image generation. CoRR, abs/1905.10485, 2019. 2 · arXiv
  48. [79]Fisher Yu, Yinda Zhang, Shuran Song, Ari Seff, and Jianxiong Xiao. Lsun: Construction of a large-scale image dataset using deep learning with humans in the loop. arXiv preprint arXiv:1506.03365, 2015. 5 · arXiv
  49. [80]Pan Zhang, Bo Zhang, Dong Chen, Lu Yuan, and Fang Wen. Cross-Domain Correspondence Learning for Exemplar-Based Image Translation. In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR, 2020. 2
  50. [81]Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shechtman, and Oliver Wang. The Unreasonable Effectiveness of Deep Features as a Perceptual Metric. In CVPR, 2018. 4, 11, 13
  51. [82]Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shechtman, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR, 2018. 45, 46
  52. [83]Bolei Zhou, Hang Zhao, Xavier Puig, Sanja Fidler, Adela Barriuso, and Antonio Torralba. Semantic understanding of scenes through the ade20k dataset. arXiv preprint arXiv:1608.05442, 2016. 6 · arXiv

Cited by

Sorted by citations

Authors
Esser, P., Rombach, R., & Ommer, B.
Affiliations
Heidelberg University (DE)
ORCID
Rombach, Robin
Venue
2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Publisher
IEEE
Published
2021-06
Pages
12868-12878
ISSN
1063-6919 (ISSN-L)
DOI
10.1109/cvpr46437.2021.01268
OpenAlex
W3111551570
Cited by
2,514
References
8
Citations by year
2026: 363 · 2025: 678 · 2024: 696 · 2023: 491 · 2022: 214 · 2021: 72
FWCI
120.44
Type
conference paper
Language
en
Open access
Licence
Locations
  • IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University)
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Generative Adversarial Networks and Image Synthesis
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications
Keywords
high-resolution image generation, transformer models, convolutional neural networks, CNN-Transformer hybrid, long-range interactions, inductive bias, conditional image generation, semantic segmentation guidance

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review