The DOI matches its Crossref record; 7 of the 7 references it cites were found in a registry or an index.
Diverse Image Generation via Self-Conditioned GANs
Open accessAbstract
We introduce a simple but effective unsupervised method for generating diverse images. We train a class-conditional GAN model without using manually annotated class labels. Instead, our model is conditional on labels automatically derived from clustering in the discriminator’s feature space. Our clustering step automatically discovers diverse modes, and explicitly requires the generator to cover them. Experiments on standard mode collapse benchmarks show that our method outperforms several competing methods when addressing mode collapse. Our method also performs well on large-scale datasets such as ImageNet and Places365, improving both diversity and standard metrics (e.g., Fréchet Inception Distance), compared to previous methods.
Overview
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.
Integrity signals
Registry
Cites
Statistics
Verification
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.
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?
Representational convergence could explain why conditional generation is easier than unconditional generation.
“Our hypothesis could explain the phenomenon that conditional generation is easier than unconditional (Mirza & Osindero, 2014;Liu et al., 2020;Sauer et al., 2022), as the data we condition on may have the same platonic structure as the data we are generating.”
From The Platonic Representation Hypothesis (2024) · cites this paper as Liu et al., 2020
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
The Hungarian method for the assignment problem
Kuhn, H. W. (1955) · Naval Research Logistics Quarterly · doi:10.1002/nav.3800020109 · Cited by 13,120
A Style-Based Generator Architecture for Generative Adversarial Networks
Karras, T., Laine, S., & Aila, T. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00453 · Cited by 11,714
Conditional Generative Adversarial Nets
Mirza, M., & Osindero, S. (2014) · arXiv · doi:10.48550/arxiv.1411.1784 · Cited by 8,863
Improved training of Wasserstein GANs
Gulrajani, I., Ahmed, F., Arjovsky, M., Dumoulin, V., & Courville, A. (2017) · arXiv · doi:10.48550/arxiv.1704.00028 · Cited by 6,800
Large Scale GAN Training for High Fidelity Natural Image Synthesis
Brock, A., Donahue, J., & Simonyan, K. (2018) · arXiv · doi:10.48550/arxiv.1809.11096 · Cited by 3,717
Flagged first, then by citations
Research works printed but not linked: 38
- tenenbaum, 2000, A global geometric framework for nonlinear dimensionality reduction, Science, 290, 2319, 10.1126/science.290.5500.2319 · DOI · Paper page
- akash, 2017, Veegan: Reducing mode collapse in gans using implicit variational learning, NeurIPS
- narayanan, 2010, Sample complexity of testing the manifold hypothesis, Advances in neural information processing systems, 1786
- mukherjee, 2018, ClusterGAN Latent space clustering in generative adversarial networks, 2
- miyato, 2018, Spectral normalization for generative adversarial networks, ICLRE, 4
- salimans, 2016, Improved techniques for training GANs, NeurIPS, 1
- 10.1109/CVPR.2018.00616 · DOI
- radford, 2016, Unsupervised representation learning with deep convolutional generative adversarial networks, ICLRE, 2
- paszke, 2019, Pytorch torchvision models, 6
- arnab, 2018, Multi-agent diverse generative adversarial networks, CVPR, 2
- 10.1007/978-3-319-46493-0_20 · DOI
- gidaris, 2018, Unsupervised representation learning by predicting image rotations, ICLRE, 2
- goodfellow, 2016, Nips 2016 tutorial Generative adversarial networks, 1
- goodfellow, 2014, Generative adversarial nets, NIPS
- hoang, 2018, Mgan: Training generative adversarial nets with multiple generators, ICLRE
- hoffman, 2018, Cycada: Cycle-consistent adversarial domain adaptation, ICML, 2
- huang, 2018, Multimodal unsupervised image-to-image translation, ECCV, 2
- 10.1109/CVPR.2017.632 · DOI · Paper page
- karras, 2018, Progressive growing of gans for improved quality, stability, and variation, ICLRE, 2
- mescheder, 2018, Which training methods for gans do actually converge?, ICML, 2
- bottou, 1995, Convergence properties of the k-means algorithms, NeurIPS, 4
- xudong, 2017, Least squares generative adversarial networks, CVPR, 2
- 10.1109/ICCV.2019.00460 · DOI
- andrew, 2017, Neural photo editing with introspective adversarial networks, ICLRE, 6
- metz, 2017, Unrolled Generative Adversarial Networks, ICLRE, 2
- denton, 2015, Deep generative image models using a laplacian pyramid of adversarial networks, NeurIPS, 2
- arthur, 2007, k-means-+: The advantages of careful seeding, Proceedings of the Eighteenth Annual ACM-SIAM Symposium on Discrete Algorithms, 1027
- 10.1109/CVPR.2019.00597 · DOI
- martín arjovsky, 2017, Wasser-stein generative adversarial networks, ICML, 2
- 10.1109/ICCV.2017.244 · DOI · Paper page
- richard, 2018, The unreasonable effectiveness of deep features as a perceptual metric, CVPR, 7
- lecun, 1998, The MNIST Database of Handwritten Digits, 5
- zhang, 2017, Stack-gan: Text to photo-realistic image synthesis with stacked generative adversarial networks, ICCV, 2
- 10.1109/ICCV.2019.00828 · DOI
- zhu, 2016, Generative visual manipulation on the natural image manifold, ECCV, 6
- lucic, 2019, High-fidelity image generation with fewer labels, ICML, 1
- zhou, 2014, Learning deep features for scene recognition using places database, NeurIPS, 4
- lin, 2018, Pacgan: The power of two samples in generative adversarial networks, NeurIPS
Cited by
Sorted by citations
The Platonic Representation Hypothesis
Huh, M., Cheung, B., Wang, T., & Isola, P. (2024) · arXiv · doi:10.48550/arxiv.2405.07987 · Cited by 35
“Our hypothesis could explain the phenomenon that conditional generation is easier than unconditional (Mirza & Osindero, 2014; Liu et al., 2020; Sauer et al., 2022), as the data we condition on may have the same platonic structure as the data we are generating.”
BackgroundSelf-citation
Metadata
- Authors
- Liu, S., Wang, T., Bau, D., Zhu, J.-Y., & Torralba, A.
- Affiliations
- MIT Computer Science and Artificial Intelligence Laboratory (US)
- Adobe Systems (United States) (US)
- Venue
- 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2020-06
- Pages
- 14274-14283
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W3034523045
- Cited by
- 91
- References
- 7
- Citations by year
- 2026: 3 · 2025: 9 · 2024: 19 · 2023: 19 · 2022: 14 · 2021: 24 · 2020: 3
- FWCI
- 4.63
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- 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)
- DSpace@MIT (Massachusetts Institute of Technology) (submittedVersion, cc-by-nc-sa)
- arXiv (Cornell University) (cc-by)
- Funders
- National Science Foundation (1524817, 1447476)
- Nvidia
- Defense Advanced Research Projects Agency (FA8750, FA8750-18-C000)
- 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 › Digital Media Forensic Detection
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Keywords
- mode collapse, image generation, Fréchet Inception Distance, generative adversarial networks
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
