The DOI matches its Crossref record; 42 of the 43 references it cites were found in a registry or an index.
Effectively Unbiased FID and Inception Score and Where to Find Them
Open accessAbstract
This paper shows that two commonly used evaluation metrics for generative models, the Fréchet Inception Distance (FID) and the Inception Score (IS), are biased -- the expected value of the score computed for a finite sample set is not the true value of the score. Worse, the paper shows that the bias term depends on the particular model being evaluated, so model A may get a better score than model B simply because model A's bias term is smaller. This effect cannot be fixed by evaluating at a fixed number of samples. This means all comparisons using FID or IS as currently computed are unreliable. We then show how to extrapolate the score to obtain an effectively bias-free estimate of scores computed with an infinite number of samples, which we term FID Infinity and IS Infinity. In turn, this effectively bias-free estimate requires good estimates of scores with a finite number of samples. We show that using Quasi-Monte Carlo integration notably improves estimates of FID and IS for finite sample sets. Our extrapolated scores are simple, drop-in replacements for the finite sample scores. Additionally, we show that using low discrepancy sequence in GAN training offers small improvements in the resulting generator.
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?
In the citing paper
“FID is crucial for gauging the similarity in style and content between two image sets, where a lower FID score signifies greater resemblance.”
From ChatScratch: An AI-Augmented System Toward Autonomous Visual Programming Learning for Children Aged 6-12 (2024) · cites this paper as [11]
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
Rethinking the Inception Architecture for Computer Vision
Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., & Wojna, Z. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.308 · Cited by 32,138
Learning Multiple Layers of Features from Tiny Images
Krizhevsky, A. (2024) · doi:10.57702/zp44cu3g · Cited by 23,889
Unpaired image-to-image translation using cycle-consistent adversarial networks
Zhu, J.-Y., Park, T., Isola, P., & Efros, A. A. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.244 · Cited by 23,602
GAN（Generative Adversarial Nets）
淳司, 柴. (2017) · Journal of Japan Society for Fuzzy Theory and Intelligent Informatics · doi:10.3156/jsoft.29.5_177_2 · Cited by 20,507
Auto-Encoding Variational Bayes
Kingma, D. P., & Welling, M. (2013) · arXiv · doi:10.48550/arxiv.1312.6114 · Cited by 20,067
Unsupervised representation learning with deep convolutional generative adversarial networks
Radford, A., Metz, L., & Chintala, S. (2015) · arXiv · doi:10.48550/arxiv.1511.06434 · Cited by 12,235
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
GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
Heusel, M., Ramsauer, H., Unterthiner, T., Nessler, B., & Hochreiter, S. (2017) · arXiv · doi:10.48550/arxiv.1706.08500 · Cited by 8,365
Flagged first, then by citations
1 of the 43 references couldn’t be matched to a known paper.
Research works printed but not linked: 4
- smolyak, 1963, Quadrature and interpolation formulas for tensor products of certain classes of functions, Doklady Akademii Nauk, 148, 1042
- goodfellow, 0, Generative adversarial nets, Advances in neural information processing systems, 2672
- owen, 0, Quasi-monte carlo sampling
- zhou, 2019, Hype Human eye perceptual evaluation of generative models
Cited by
Sorted by citations
Items 201–242 of 242
Pixelating to the Edge: Generative AI Art on Edge Devices
Pardesi, S. D., Muntean, C. H., & Simiscuka, A. A. (2025) · IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB) · doi:10.1109/bmsb65076.2025.11165718
Polenghi, M., Caldelli, R., Loconsole, C., & Ogliari, E. (2025) · IEEE International Conference on Environment and Electrical Engineering and IEEE Industrial and Commercial Power Systems Europe (EEEIC / I&CPS Europe) · doi:10.1109/eeeic/icpseurope64998.2025.11169106
Gao, Z., Zhao, L., Zheng, J., Tao, Y., Zhang, C., & Wang, Q. (2025) · Scientific Reports · doi:10.1038/s41598-025-20962-2
Long Range Constraints for Neural Texture Synthesis Using Sliced Wasserstein Loss
Yin, L., & Chua, A. (2025) · Machine learning for signal processing ... · doi:10.1109/mlsp62443.2025.11204346
Domain-Agnostic Neural Oil Painting via Normalization Affine Test-Time Adaptation
Dong, Q., Liu, L., Wang, Y., Liu, J. J. R., & Zheng, Z. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3758165 · Cited by 3
Kennedy, A., & Everson, R. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227282
Generative Artificial Intelligence
Shrivastav, A., Sankathala, S., Chaturvedi, K., Patre, K., Das, S., & Shaikh, R. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-5616-7.ch002
Preserving instance-level characteristics for multi-instance generation
Ryu, J., Moon, S.-W., & Cho, D. (2025) · Image and Vision Computing · doi:10.1016/j.imavis.2025.105851
BananaSense: A Visual Analytics and Evaluation Method for Fruit Maturity-Aware Image Generation
Gao, Y., & Chen, Y. (2026) · Communications in computer and information science · doi:10.1007/978-981-95-4966-5_6
The Fréchet Coefficient for GANs and Diffusion Models Evaluation
Kucharski, A., & Fabijańska, A. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-04197-5_14
Hung, J.-C., Pan, S.-P., Wu, J., Jaw, D.-W., & Huang, S.-C. (2026) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2026.3651222
Qiao, T. (2025) · International Conference on Image Processing, Computer Vision and Machine Learning (ICICML) · doi:10.1109/icicml67980.2025.11333452
FLD+: Data-Efficient Evaluation Metric for Generative Models
Pranav Jeevan, P., Nixon, N., & Sethi, A. (2025) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw69036.2025.00175
Methodology for Real-Time Privacy Preserving Data Recording on Autonomous Vehicles
Naranjo, R., Aranjuelo, N., & Rodríguez-Moreno, I. (2025) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc60802.2025.11423575
Improving rare-class detection in deep-sea imagery via generative augmentation with stable diffusion
Deng, J., Duan, M., Wei, D., Song, W., He, X., & Xia, J. (2026) · Scientific Reports · doi:10.1038/s41598-026-45732-6
Text-Based Semantic Image Editing
Couairon, G. (2023) · theses.fr (ABES) · doi:10.70675/c4c397d4zb2e4z41f7z9374z64ed51cfba38
Quality and Diversity in Generative Models through the lens of f-divergences
Vérine, A. (2024) · theses.fr (ABES) · doi:10.70675/36e5bb7cz8351z4ba7z8fffzcc9523acc864
Image editing with deep neural networks
Grechka, A. (2023) · theses.fr (ABES) · doi:10.70675/804c7aafzd28dz4860z951fz82e38c31b8d3
Guiding neural networks for image colorization through user interactions
Carrillo, H. (2024) · theses.fr (ABES) · doi:10.70675/02421605z6c2bz4faaza888z71d763702748
Transcriptomics data generation with deep generative models
Lacan, A. (2025) · theses.fr (ABES) · doi:10.70675/4bbb4d4fzed5ez4f41z86f4zf6f892f9e3e8
Wu, G., Zhang, X., Wu, Y., Du, H., Huang, Y., Jin, H., Tang, S., Zhang, Y., & Qian, J. (2025) · IEEE International Conference on Applied Mathematics, Modeling and Computer Simulation (AMMCS) · doi:10.1109/ammcs65761.2025.11459886
Understanding Frechet Speech Distance for Synthetic Speech Quality Evaluation
Kim, J.-W., Agarwal, D., & Cerina, F. (2026) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp55912.2026.11464031
Bai, J., Liu, H. D., Su, Y., & Geng, H. (2026) · Springer eBooks · doi:10.1007/978-981-95-0836-5_14
Elsharif, W., Alzubaidi, M. S., She, J. P. M., & Agus, M. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-22267-1_26
Lu, Y., Miller, B., Pleis, C., & Ardiles-Cruz, E. (2026) · International Conference on Smart Applications, Communications and Networking (SmartNets) · doi:10.1109/smartnets69662.2026.11604703
ContextDiff: Context-aware controllable generation with text-to-image diffusion models
Djenouri, Y., Belbachir, A. N., Belhadi, A., & Srivastava, G. (2026) · Journal of Intelligent & Fuzzy Systems · doi:10.1177/18758967261469967
Are Wavelets More Reliable? Comparing the Topology of Image Generator Metrics
Patil, M., Tiwari, A., Bharadwaj, A., & Gadre, V. M. (2026) · International Conference on Signal Processing and Communications (SPCOM) · doi:10.1109/spcom71105.2026.11623005
Fréchet Wavelet Style Distance: an Interpretable Image Style Similarity Metric
Bharadwaj, A., Kumar, A., Kumar, D., & Gadre, V. M. (2026) · Proceedings - International Conference on Image Processing · doi:10.1109/icip61757.2026.11630062
SSGLD: Spatial-Semantic Guided Latent Diffusion for Person Image Synthesis
Xu, Y., Zhang, Z., & Chen, G. (2026) · IEEE Access · doi:10.1109/access.2026.3726280
PicTune: A Framework for Optimal Selection on Text-to-Image Diffusion Models Fine-Tuning
Lewandowski, B., Birke, R., & Chen, L. Y. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37657-2_25
Fallah, S., & Herzog, N. J. (2026) · Journal of Imaging · doi:10.3390/jimaging12090426
Huang, J., He, J., Chen, Z., Shen, L., He, L., & Gui, W. (2026) · Advanced Engineering Informatics · doi:10.1016/j.aei.2026.105282
Variational autoencoder in analysis of motion capture data
Furs, M., Hanak, D., & Świtoński, A. (2026) · Neural Computing and Applications · doi:10.1007/s00521-026-12469-2
THEval. Evaluation Framework for Talking Head Video Generation
Quignon, N., Chopin, B., Wang, Y., & Dantcheva, A. (2026) · doi:10.1109/cvprf71909.2026.00191
Assessing the Reliability of Image Quality Metrics and Mitigating Quality Bias in Generative Models
Jung, H., Lu, S., De, W., & Wang, X. (2026) · doi:10.1109/cvprf71909.2026.00764
Nagy, R., Voß, H., Hoang-Minh, T., Tsakov, M., Nikolov, T., Zhang, Z., Ao, T., Yang, S., Huang, S., Cheng, Y., et al. (2026) · doi:10.1109/cvprf71909.2026.00211
Diffusion-Based Inference for Heterogeneous Multichannel Fluorescence ImagingPreprint
Georvasilis, I., Frías, M., Badosa, C., Musokhranova, U., Oyarzábal, A., Jimenez-Mallebrera, C., Rumiński, J., & Roldán, M. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.10.07.753438
Cao, H., & Chen, J. (2026) · Mathematics · doi:10.3390/math14203759
Desai, V., Bench, C., Roozemond, C., van Engen, R. V., & Thomas, S. A. (2026) · IEEE International Joint Conference on Neural Network · doi:10.1109/ijcnn69854.2026.11728785
IGAN: An Inception-Based Generative Adversarial Network for Multi-Scale Image Synthesis
Hashim, A. A., Al-Shuwaili, A., Hashim, A. A., Al-Bayaty, A., & Alzubaidi, A. H. (2026) · Journal of Imaging · doi:10.3390/jimaging12100518
Chen, J., Wang, J., Li, X., Cheng, R., Wei, W., Yang, Y., Wang, W., & Qiu, S. (2026) · Applied Soft Computing · doi:10.1016/j.asoc.2026.116758
Metadata
- Authors
- Chong, M. J., & Forsyth, D.
- Affiliations
- University of Illinois Urbana-Champaign (US)
- Venue
- 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2020-06
- Pages
- 6069-6078
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W2985516053
- Cited by
- 242
- References
- 43
- Citations by year
- 2026: 26 · 2025: 59 · 2024: 69 · 2023: 43 · 2022: 24 · 2021: 18 · 2020: 3
- FWCI
- 6.78
- 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)
- 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 › Digital Media Forensic Detection
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image and Video Quality Assessment
- Keywords
- Fréchet Inception Distance, Inception Score, generative adversarial networks, quasi-Monte Carlo integration, low-discrepancy sequences, GAN training, generative model evaluation
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
