Pith. sign in

REVIEW 5 major objections 6 minor 71 references

Multi-Armed Bandit Approach for Optimizing Training on Synthetic Data

T0 review · 5 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A UCB-based training loop that switches between two per-image synthetic-data quality scores improves fine-tuning accuracy by up to 10 percentage points across six architectures and three classification datasets.

desk verdict A plausible UCB-plus-metric idea with public data and code, but the core claim is contradicted by the paper's own ablation and the main metric is undefined; worth a revision cycle, not acceptance. read the letter →

arxiv 2412.05466 v1 pith:JRDVFRT4 submitted 2024-12-06 cs.LG cs.CV

Abdulrahman Kerim , Leandro Soriano Marcolino , Erickson R. Nascimento , Richard Jiang This is my paper · ORCID
classification cs.LGcs.CV
keywords syntheticdatamulti-armedbanditUCBusabilitymetricimageclassificationfine-tuningStableDiffusionfeaturecohesion
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that a training-time selection rule, not just a static quality score, determines how much synthetic images help a supervised classifier. It proposes a usability metric U with two components: one measuring diversity and photorealism from mid-level features, and one measuring feature cohesion between a synthetic image and its real class mean. A multi-armed bandit then switches between the two score-based data subsets during fine-tuning, treating validation accuracy as reward. The authors report that this dynamic selection beats static metrics such as FID, IS, SSIM, and PSNR, improving fine-tuning accuracy by up to 10 percentage points across six architectures on three classification datasets. If true, the practical lesson is that the same synthetic pool can be used more effectively by letting the model's current state decide which examples to train on.

What carries the argument

The central object is the usability score U = (Ψ, Φ) together with the UCB-based training procedure. Ψ (Diversity and Photorealism Score) is computed from the mean and standard deviation of Inception-V3 feature vectors of the image, its class, and the dataset excluding its class; Φ (Feature Cohesion Score) is computed as 1/D_KL of the normalized VGG16 features of the synthetic image against the normalized mean features of real images in the same class. The UCB procedure treats the two score components as arms, computes rewards from validation accuracy after each epoch, and switches data subsets once validation accuracy is flat for more than a patience threshold, with exploration term β√(log(total counts)/(loader counts+ε)). This mechanism is what makes the selection dynamic rather than a one-time ranking.

What would settle it

Run the UCB training procedure with the two arms' identities hidden: keep the same data subsets but label them randomly, or swap Ψ and Φ between arms. If the reported accuracy gains persist under random or swapped labels, the selection criterion itself is not the cause. Alternatively, evaluate the same pipeline on a task whose images are far from ImageNet (e.g., medical scans or satellite imagery); if the gains vanish, the ImageNet-pretrained features are the real load-bearing part.

Watch

Extended reading notes

Core claim

The paper's central claim is that per-image synthetic data usability is best captured by a two-dimensional score, U = (Ψ, Φ), where Ψ (DPS) combines the mean and standard deviation of Inception-V3 features for an image, its class, and the rest of the dataset to measure diversity and photorealism, and Φ (FCS) is the inverse KL divergence between the image's normalized VGG16 features and the normalized mean features of real images in the same class. On top of this score, the paper adds a UCB-based training loop with two arms, namely the top-M synthetic images ranked by Ψ and the top-M ranked by Φ, and switches arms when validation accuracy stops improving. The authors report that this combination achieves the best fine-tuning accuracy in Table 4 for AlexNet, EfficientNet, ViT, SwinTransformer, VGG, and REGNet on photorealistic synthetic versions of Car Accidents, CIFAR-10, and Birds-525, outperforming SSIM, PSNR, IS, and FID-based selection.

Load-bearing premise

The load-bearing premise is that two static per-image scores, computed once with ImageNet-pretrained Inception-V3 and VGG16 features, stay the right selection criterion for every downstream classifier at every epoch, and that validation accuracy can safely guide the switch between them.

Editorial extensions

If this is right

  • Fine-tuning on a fixed top-M selection of synthetic images is suboptimal; allowing the training loop to alternate between two quality criteria can raise accuracy by several points on standard classification benchmarks.
  • The reported gains hold across very different architectures (convolutional, transformer, and scaled networks), suggesting the dynamic selection helps the training procedure itself rather than one model family.
  • Traditional generative-quality metrics such as FID and IS, which score a whole dataset, can be outperformed by per-image scores that combine mid-level and high-level features.
  • Because the method only needs two precomputed feature sets, it can be added to an existing fine-tuning pipeline with modest overhead.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural extension is to apply the same two-arm UCB idea with other paired scoring criteria (for example, sharpness versus diversity, or adversarial robustness versus clean accuracy); the paper does not test this generality.
  • The static scores are computed once with ImageNet-pretrained extractors, so the approach may transfer poorly to domains far from ImageNet; testing on medical or satellite imagery would clarify whether the claim depends on those features.
  • Because the bandit reward is validation accuracy, there is a risk of overfitting the validation set when the patience and β hyperparameters are tuned on it; a cross-validated reward would make the improvement claim more defensible.
  • The generation pipeline couples an LLM with Stable Diffusion, but the paper does not compare this attribute-aware pipeline against random prompting; such a comparison would isolate the contribution of the LLM-generated attributes.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. This paper proposes a multi-armed bandit (UCB) training procedure for fine-tuning classifiers on synthetic data. The authors introduce a two-component usability score U = [Ψ, Φ], where Ψ (DPS) is intended to capture diversity and photorealism using Inception-V3 features, and Φ (FCS) measures feature cohesion via KL divergence in VGG16 feature space. A UCB algorithm is used to switch between two subsets of synthetic images selected by the two scores, based on validation accuracy. The authors also present an LLM-plus-Stable-Diffusion pipeline for generating attribute-aware synthetic datasets and release six such datasets. Experiments fine-tune six architectures on these datasets and report that the proposed U score with UCB achieves the best classification accuracy compared to SSIM, PSNR, IS, and FID.

Significance. If the proposed dynamic selection approach worked as claimed, it would address a real need in the synthetic-data literature: most existing metrics assess generation quality rather than training usability. The idea of adapting sample selection during training via a bandit is plausible, and the public release of the generated datasets is a useful community resource. However, the paper does not yet establish the claim. The DPS component is never given a closed-form definition, the experiments appear to be single runs without statistical significance, and the comparison is confounded between metric choice and the UCB mechanism. The ablation study also contains configurations where the proposed method is outperformed by UCB with traditional metrics. Thus the significance remains potential rather than demonstrated.

major comments (5)
  1. [Section 3.3, Figure 1] The DPS component Ψ is never given a closed-form definition. The text describes computing means and standard deviations of Inception-V3 features but does not specify how these are combined into Ψ; Figure 1 is a schematic with undefined symbols. Since U = [Ψ, Φ] is the central construct of the paper, the method is not reproducible without the exact formula. Please provide the precise equation for Ψ, including any normalizations and distance functions.
  2. [Section 4.5, Table 5] The ablation contains a direct counterexample to the central claim. For SP-Car-2/AlexNet, the row 'SSIM, PSNR, IS, & FID 4 Arms' reports 86.0, while 'Our Approach (U) 2 Arms' reports 80.0; VGG shows a similar inversion (80 vs 77). Since all accuracy numbers in Tables 4 and 5 are single integers with no error bars or significance tests, the statements in Section 4.3 that 'Our metric U ... achieves the best results' and 'significantly outperformed' are not supported. Please provide repeated runs with standard errors and statistical tests, or revise the claim to exclude configurations where the method is not superior.
  3. [Section 4.3, Table 4] The comparison is confounded because the proposed method uses UCB-based dynamic switching between two arms, while the baselines (SSIM, PSNR, IS, FID) appear to use static ranking and training on a single selected subset. To attribute the improvement to the U metric rather than to the bandit mechanism, the baselines should be evaluated under the same UCB procedure (e.g., each metric as a single arm, or UCB over the four metrics), or the proposed metric should also be evaluated as a static selector. As presented, the gains in Table 4 could be driven entirely by the adaptive switching.
  4. [Section 3.4, Algorithm 2] The UCB reward is the validation accuracy, but the paper does not state whether the final reported accuracy is computed on a held-out test set that is disjoint from the validation split used for arm selection. If the same split is used for both, the reported numbers can overfit the validation criterion and the evaluation is circular. Please clarify the train/validation/test partition and ensure that Table 4 reports test accuracy on a set never used for reward updates or model selection.
  5. [Section 4.1, Algorithm 2] The hyperparameters M (the number of selected synthetic images), K (the number of real images used to compute the class mean in Eq. (4)), and the patience threshold in Algorithm 2 are never given numerical values. These are essential for reproducibility. Please report the exact values used in the experiments and explain how they were chosen.
minor comments (6)
  1. [Abstract and Data Availability] The word 'publically' should be 'publicly'.
  2. [Section 3.1, Eq. (3)] The entropy term in Eq. (2) and its use in the joint objective Eq. (3) are not clearly connected; the paper does not specify how α(t) is updated over time, despite the claim that it is temporally adaptive.
  3. [Figure 1] The symbols ψ, φ, τ, υ, γ, δ appear in Figure 1 but are not defined in the text; please add a legend or name them in Section 3.3.
  4. [Table 3] The AUC values are reported without any variance or error bars, so it is unclear whether the differences between artistic, photorealistic, and real datasets are statistically meaningful.
  5. [Section 4.1] The fine-tuning setup is under-specified: the number of epochs, learning rate, batch size, optimizer, and data augmentation for each architecture are not provided.
  6. [Abstract] The phrase 'improvement of up to 10% in classification accuracy' is ambiguous: it is not clear whether this refers to absolute percentage points (e.g., from 70% to 80%) or relative improvement (e.g., from 70% to 77%).

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the U metric and UCB procedure are evaluated against held-out test accuracy, and the central claim is not constructed from its own output.

full rationale

The paper's derivation chain is empirically self-contained rather than circular. Synthetic images are produced by an LLM+Stable Diffusion pipeline (Algorithm 1); per-image usability scores U = (Psi, Phi) are computed by fixed, pretrained ImageNet feature extractors (Inception-V3 and VGG16) using statistics and KL divergences against real and synthetic feature distributions (Section 3.3); top-M images are filtered; and the UCB procedure in Algorithm 2 selects between the two score-based loaders using validation accuracy as the reward. The reported claim is test classification accuracy on held-out real data, which is not an input to the metric definitions or to the UCB reward computation. The UCB mechanism is a standard validation-guided model/arm selection scheme, and selecting the better-scoring arm by validation accuracy does not make the downstream test accuracy a definitional consequence. No fitted parameter is renamed as a prediction: Psi and Phi are not fit to the reported accuracies, and the UCB hyperparameters are fixed (epsilon = 1e-5, beta = 2). The self-citations present (the first author's book and prior WACV paper) are contextual and not load-bearing for the central claim, and no uniqueness theorem or ansatz is imported from the authors' prior work. The internal tension in Table 5, where the 4-arm SSIM/PSNR/IS/FID configuration beats the proposed 2-arm U on some cells (e.g., 86.0 vs 80.0 for AlexNet on SP-Car-2), is a matter of evidence consistency and reproducibility, not circularity: the result is still an independent empirical outcome, even if it undercuts the paper's headline assertion. Therefore no circular step can be exhibited, and the appropriate score is 0.

Assumptions & free parameters 8 free parameters · 5 assumptions · 0 invented entities

The central method rests on several unreported free parameters (alpha(t), patience, M, K, normalization) and on two strong domain assumptions: that ImageNet-pretrained features measure synthetic-data usability, and that validation accuracy is a safe reward for choosing data subsets. These are not explicitly fitted to the final test numbers, but because the exact values are omitted and ablations show competing configurations performing differently, the reader cannot separate the method's contribution from the chosen configuration. No new physical entities are introduced.

free parameters (8)
  • alpha(t) in Eq. 3 = not specified
    Balances photorealism and diversity in the joint objective, but no schedule or value is given and it does not appear in the final U score.
  • beta (UCB exploration constant) = 2
    Set in Algorithm 2 without sensitivity analysis; affects how often the training procedure switches arms.
  • epsilon (UCB denominator) = 1e-5
    Chosen to avoid division by zero; not varied across experiments.
  • patience threshold = not reported
    Controls how many epochs without validation improvement trigger an arm switch; no value is stated.
  • M (number of selected synthetic images) = half the dataset (from Table 4 caption)
    The paper says half of the synthetic datasets are used for fine-tuning, but the exact selection count and whether M is tuned are not specified.
  • K (number of real images per class for class mean) = not reported
    K controls the averaged real class features used in the FCS computation; no value or sampling strategy is given.
  • DPS internal statistics = not given
    The Diversity and Photorealism Score (Psi) is described in a figure but no closed-form equation is provided in the text.
  • normalization for FCS = not specified
    The KL divergence in Eq. 4 requires normalized feature vectors, but the normalization method (softmax, L2, etc.) is not stated.
assumptions (5)
  • domain assumption ImageNet-pretrained Inception-V3 and VGG16 features are reliable proxies for photorealism, diversity, and class-level cohesion for arbitrary downstream classification tasks.
    Used throughout Section 3.3 to define DPS and FCS; no evidence is provided that these features transfer to the three tested tasks.
  • domain assumption KL divergence between normalized feature vectors is a valid per-image cohesion score.
    Eq. 4 defines FCS as the inverse of a KL divergence, which presumes the normalized features behave like probability distributions over the same support.
  • domain assumption Validation accuracy is an unbiased reward for choosing between synthetic data subsets.
    Algorithm 2 uses validation accuracy to update UCB rewards; if the same validation set is used for final model selection, the reported test accuracies may be optimistically biased.
  • domain assumption Per-image usability is separable and can be ranked before training.
    The method computes all Phi and Psi scores once before fine-tuning and then selects top images, which ignores the model's evolving state beyond the two-arm switching.
  • standard math The UCB exploration-exploitation framework applies to data loader selection.
    The paper applies a standard UCB formula to arm selection; this is background material, not derived in the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Multi-Armed Bandit Approach for Optimizing Training on Synthetic Data." pith.science (2026). https://pith.science/paper/JRDVFRT4

@misc{pith2026241205466,
  author       = {Pith},
  title        = {Pith review of: Multi-Armed Bandit Approach for Optimizing Training on Synthetic Data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JRDVFRT4}},
  note         = {Machine review of arXiv:2412.05466}
}
read the original abstract

Supervised machine learning methods require large-scale training datasets to perform well in practice. Synthetic data has been showing great progress recently and has been used as a complement to real data. However, there is yet a great urge to assess the usability of synthetically generated data. To this end, we propose a novel UCB-based training procedure combined with a dynamic usability metric. Our proposed metric integrates low-level and high-level information from synthetic images and their corresponding real and synthetic datasets, surpassing existing traditional metrics. By utilizing a UCB-based dynamic approach ensures continual enhancement of model learning. Unlike other approaches, our method effectively adapts to changes in the machine learning model's state and considers the evolving utility of training samples during the training process. We show that our metric is an effective way to rank synthetic images based on their usability. Furthermore, we propose a new attribute-aware bandit pipeline for generating synthetic data by integrating a Large Language Model with Stable Diffusion. Quantitative results show that our approach can boost the performance of a wide range of supervised classifiers. Notably, we observed an improvement of up to 10% in classification accuracy compared to traditional approaches, demonstrating the effectiveness of our approach. Our source code, datasets, and additional materials are publically available at https://github.com/A-Kerim/Synthetic-Data-Usability-2024.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

71 extracted references · 59 canonical work pages

  1. [1]

    International Journal of Computer Vision, 1–25 (2024)

    Mettes, P., Ghadimi Atigh, M., Keller-Ressel, M., Gu, J., Yeung, S.: Hyperbolic deep learning in computer vision: A survey. International Journal of Computer Vision, 1–25 (2024)

    work page 2024

  2. [2]

    IEEE Transactions on Artificial Intelligence (2024)

    Dubey, S.R., Singh, S.K.: Transformer-based generative adversarial networks in computer vision: A comprehensive survey. IEEE Transactions on Artificial Intelligence (2024)

    work page 2024

  3. [3]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Xu, J., De Mello, S., Liu, S., Byeon, W., Breuel, T., Kautz, J., Wang, X.: Groupvit: Semantic segmentation emerges from text supervision. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 18134– 18144 (2022)

    work page 2022

  4. [4]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Junayed, M.S., Sadeghzadeh, A., Islam, M.B., Wong, L.-K., Aydın, T.: HiMODE: A Hybrid Monocular Omnidirectional Depth Estimation Model. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 5212–5221 (2022)

    work page 2022

  5. [5]

    : Internimage: Exploring large-scale vision foundation models with deformable convolutions

    Wang, W., Dai, J., Chen, Z., Huang, Z., Li, Z., Zhu, X., Hu, X., Lu, T., Lu, L., Li, H., et al. : Internimage: Exploring large-scale vision foundation models with deformable convolutions. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 14408–14419 (2023)

    work page 2023

  6. [6]

    In: European Conference on Computer Vision (2014)

    Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Doll´ ar, P., Zitnick, C.L.: Microsoft COCO: Common Objects in Context. In: European Conference on Computer Vision (2014)

    work page 2014

  7. [7]

    International Journal of Computer Vision 127, 302–321 (2019) 20

    Zhou, B., Zhao, H., Puig, X., Xiao, T., Fidler, S., Barriuso, A., Torralba, A. International Journal of Computer Vision 127, 302–321 (2019) 20

    work page 2019

  8. [8]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp

    Zhou, B., Zhao, H., Puig, X., Fidler, S., Barriuso, A., Torralba, A.: Scene parsing through ade20k dataset. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 633–641 (2017)

    work page 2017

Show all 71 references
  1. [9]

    Krizhevsky, A., Hinton, G., et al.: Learning multiple layers of features from tiny images (2009)

    2009

  2. [10]

    In: 2009 IEEE Conference on Computer Vision and Pattern Recognition (2009)

    Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., Fei-Fei, L.: ImageNet: A Large- Scale Hierarchical Image Database. In: 2009 IEEE Conference on Computer Vision and Pattern Recognition (2009)

    2009

  3. [11]

    : Panda-70m: Captioning 70m videos with multiple cross-modality teachers

    Chen, T.-S., Siarohin, A., Menapace, W., Deyneka, E., Chao, H.-w., Jeon, B.E., Fang, Y., Lee, H.-Y., Ren, J., Yang, M.-H., et al. : Panda-70m: Captioning 70m videos with multiple cross-modality teachers. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern...

    2024

  4. [12]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Chen, H., Hou, Y., Qu, C., Testini, I., Hong, X., Jiao, J.: 360+ x: A panoptic multi- modal scene understanding dataset. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 19373–19382 (2024)

    2024

  5. [13]

    Packt Publishing, Birmingham, UK (2023)

    Kerim, A.: Synthetic Data for Machine Learning: Revolutionize Your Approach to Machine Learning with this Comprehensive Conceptual Guide. Packt Publishing, Birmingham, UK (2023). https://books.google.co.uk/books?id=JpXeEAAAQBAJ

    2023

  6. [14]

    International Journal of Computer Vision, 1–37 (2024)

    Delussu, R., Putzu, L., Fumera, G.: Synthetic data for video surveillance appli- cations of computer vision: A review. International Journal of Computer Vision, 1–37 (2024)

    2024

  7. [15]

    Computer Science Review 48, 100553 (2023)

    Iglesias, G., Talavera, E., D ´ ıaz-´Alvarez, A.: A survey on GANs for computer vision: Recent research, analysis and taxonomy. Computer Science Review 48, 100553 (2023)

    2023

  8. [16]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Kang, M., Zhu, J.-Y., Zhang, R., Park, J., Shechtman, E., Paris, S., Park, T.: Scaling up GANs for text-to-image synthesis. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 10124–10134 (2023)

    2023

  9. [17]

    Computational intelligence and neuroscience 2020 (2020)

    Jin, L., Tan, F., Jiang, S., et al.: Generative adversarial network technologies and applications in computer vision. Computational intelligence and neuroscience 2020 (2020)

    2020

  10. [18]

    Advances in Neural Information Processing Systems 36 (2024)

    Nguyen, Q., Vu, T., Tran, A., Nguyen, K.: Dataset diffusion: Diffusion-based syn- thetic data generation for pixel-level semantic segmentation. Advances in Neural Information Processing Systems 36 (2024)

    2024

  11. [19]

    IEEE Transactions on Pattern Analysis and Machine Intelligence45(9), 10850–10869 (2023)

    Croitoru, F.-A., Hondru, V., Ionescu, R.T., Shah, M.: Diffusion models in vision: 21 A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence45(9), 10850–10869 (2023)

    2023

  12. [20]

    In: 32nd USENIX Security Symposium (USENIX Security 23), pp

    Carlini, N., Hayes, J., Nasr, M., Jagielski, M., Sehwag, V., Tramer, F., Balle, B., Ippolito, D., Wallace, E.: Extracting training data from diffusion models. In: 32nd USENIX Security Symposium (USENIX Security 23), pp. 5253–5270 (2023)

    2023

  13. [21]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Bansal, A., Chu, H.-M., Schwarzschild, A., Sengupta, S., Goldblum, M., Geiping, J., Goldstein, T.: Universal guidance for diffusion models. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 843–852 (2023)

    2023

  14. [22]

    In: 2024 IEEE International Conference on Robotics and Automation (ICRA), pp

    Maxey, C., Choi, J., Lee, H., Manocha, D., Kwon, H.: UA V-Sim: NeRF-based Syn- thetic Data Generation for UA V-based Perception. In: 2024 IEEE International Conference on Robotics and Automation (ICRA), pp. 5323–5329 (2024). IEEE

    2024

  15. [23]

    NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis

  16. [24]

    In: Conference on Robot Learning, pp

    Dosovitskiy, A., Ros, G., Codevilla, F., Lopez, A., Koltun, V.: CARLA: An open urban driving simulator. In: Conference on Robot Learning, pp. 1–16 (2017). PMLR

    2017

  17. [25]

    In: Field and Service Robotics, pp

    Shah, S., Dey, D., Lovett, C., Kapoor, A.: Airsim: High-fidelity visual and physical simulation for autonomous vehicles. In: Field and Service Robotics, pp. 621–635 (2018). Springer

    2018

  18. [26]

    In: 2020 6th International Conference on Control, Automation and Robotics (ICCAR), pp

    Ma, C., Zhou, Y., Li, Z.: A New Simulation Environment Based on Airsim, ROS, and PX4 for Quadcopter Aircrafts. In: 2020 6th International Conference on Control, Automation and Robotics (ICCAR), pp. 486–490 (2020). IEEE

    2020

  19. [27]

    In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp

    Kerim, A., Ramos, W.L., Marcolino, L.S., Nascimento, E.R., Jiang, R.: Lever- aging synthetic data to learn video stabilization under adverse conditions. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp. 6931–6940 (2024)

    2024

  20. [28]

    Artificial intelligence review 56(9), 9221–9265 (2023)

    Paulin, G., Ivasic-Kos, M.: Review and analysis of synthetic dataset generation methods and techniques for application in computer vision. Artificial intelligence review 56(9), 9221–9265 (2023)

    2023

  21. [29]

    In: 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp

    Tobin, J., Fong, R., Ray, A., Schneider, J., Zaremba, W., Abbeel, P.: Domain randomization for transferring deep neural networks from simulation to the real world. In: 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 23–30 (2017). IEEE

    2017

  22. [30]

    Nikolenko, S.I.: Synthetic Data for Deep Learning vol. 174. Springer, ??? (2021)

    2021

  23. [31]

    IEEE Access 7, 22 64323–64350 (2019)

    Gong, Z., Zhong, P., Hu, W.: Diversity in Machine Learning. IEEE Access 7, 22 64323–64350 (2019)

    2019

  24. [32]

    Mathematical Problems in Engineering 2022 (2022)

    Katayama, T., Song, T., Jiang, X., Leu, J.-S., Shimamoto, T.: Domain adaptation through photorealistic enhanced images for semantic segmentation. Mathematical Problems in Engineering 2022 (2022)

    2022

  25. [33]

    Journal of Imaging 8(11), 310 (2022)

    Man, K., Chahl, J.: A Review of Synthetic Image Data and Its Use in Computer Vision. Journal of Imaging 8(11), 310 (2022)

    2022

  26. [34]

    arXiv preprint arXiv:2204.08610 (2022)

    Yang, S., Xiao, W., Zhang, M., Guo, S., Zhao, J., Shen, F.: Image Data Augmentation for Deep Learning: A Survey. arXiv preprint arXiv:2204.08610 (2022)

    2022 arXiv

  27. [35]

    arXiv preprint arXiv:1901.09981 (2019)

    Kariyappa, S., Qureshi, M.K.: Improving Adversarial Robustness of Ensembles with Diversity Training. arXiv preprint arXiv:1901.09981 (2019)

    2019 arXiv

  28. [36]

    Advances in neural information processing systems 30 (2017)

    Heusel, M., Ramsauer, H., Unterthiner, T., Nessler, B., Hochreiter, S.: GANs trained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems 30 (2017)

    2017

  29. [37]

    Advances in neural information processing systems 29 (2016)

    Salimans, T., Goodfellow, I., Zaremba, W., Cheung, V., Radford, A., Chen, X.: Improved techniques for training GANs. Advances in neural information processing systems 29 (2016)

    2016

  30. [38]

    In: Proceedings of the British Machine Vision Conference 2016, BMVC (2016)

    Shafaei, A., Little, J.J., Schmidt, M.: Play and learn: Using video games to train computer vision models. In: Proceedings of the British Machine Vision Conference 2016, BMVC (2016). BMV A

    2016

  31. [39]

    In: 2022 26th International Conference on Pattern Recognition (ICPR), pp

    Kiefer, B., Ott, D., Zell, A.: Leveraging synthetic data in object detection on unmanned aerial vehicles. In: 2022 26th International Conference on Pattern Recognition (ICPR), pp. 3564–3571 (2022). IEEE

    2022

  32. [40]

    In: Computer Graphics Forum, vol

    Kerim, A., Aslan, C., Celikcan, U., Erdem, E., Erdem, A.: Nova: Rendering virtual worlds with humans for computer vision tasks. In: Computer Graphics Forum, vol. 40, pp. 258–272 (2021). Wiley Online Library

    2021

  33. [41]

    Automation in Construction 155, 105060 (2023)

    Lee, H., Jeon, J., Lee, D., Park, C., Kim, J., Lee, D.: Game engine-driven synthetic data generation for computer vision-based safety monitoring of construction workers. Automation in Construction 155, 105060 (2023)

    2023

  34. [42]

    International Journal of Computer Vision129(1), 225–245 (2021)

    Wang, Q., Gao, J., Lin, W., Yuan, Y.: Pixel-wise crowd understanding via synthetic data. International Journal of Computer Vision129(1), 225–245 (2021)

    2021

  35. [43]

    In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp

    Rao, Q., Yu, X., Navasardyan, S., Shi, H.: Sim2realvs: A new benchmark for video stabilization with a strong baseline. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp. 5406–5415 (2023) 23

    2023

  36. [44]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Sun, X., Zheng, L.: Dissecting person re-identification from the viewpoint of view- point. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 608–617 (2019)

    2019

  37. [45]

    In: International Conference on Machine Learning, pp

    Cobbe, K., Hesse, C., Hilton, J., Schulman, J.: Leveraging procedural generation to benchmark reinforcement learning. In: International Conference on Machine Learning, pp. 2048–2056 (2020). PMLR

    2020

  38. [46]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Li, A., Wu, J., Liu, Y., Li, L.: Bridging the synthetic-to-authentic gap: Distortion- guided unsupervised domain adaptation for blind image quality assessment. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 28422–28431 (2024)

    2024

  39. [47]

    arXiv preprint arXiv:2401.01629 (2024)

    Hao, S., Han, W., Jiang, T., Li, Y., Wu, H., Zhong, C., Zhou, Z., Tang, H.: Synthetic data in ai: Challenges, applications, and ethical implications. arXiv preprint arXiv:2401.01629 (2024)

    2024 arXiv

  40. [48]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Ali, H., Gr¨ onlund, C., Shah, Z.: Leveraging GANs for Data Scarcity of COVID- 19: Beyond the Hype. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 659–667 (2023)

    2023

  41. [49]

    Information Sciences 586, 485–500 (2022)

    Torfi, A., Fox, E.A., Reddy, C.K.: Differentially private synthetic medical data generation using convolutional GANs. Information Sciences 586, 485–500 (2022)

    2022

  42. [50]

    Medical Image Analysis 84, 102688 (2023)

    Al Khalil, Y., Amirrajab, S., Lorenz, C., Weese, J., Pluim, J., Breeuwer, M.: On the Usability of Synthetic Data for Improving the Robustness of Deep Learning- based Segmentation of Cardiac Magnetic Resonance Images. Medical Image Analysis 84, 102688 (2023)

    2023

  43. [51]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Li, L., Lian, Q., Wang, L., Ma, N., Chen, Y.-C.: Lift3D: Synthesize 3D Training Data by Lifting 2D GAN to 3D Generative Radiance Field. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 332–341 (2023)

    2023

  44. [52]

    arXiv preprint arXiv:1705.07215 (2017)

    Kodali, N., Abernethy, J., Hays, J., Kira, Z.: On convergence and stability of GANs. arXiv preprint arXiv:1705.07215 (2017)

    2017 arXiv

  45. [53]

    Advances in neural information processing systems 30 (2017)

    Gulrajani, I., Ahmed, F., Arjovsky, M., Dumoulin, V., Courville, A.C.: Improved training of Wasserstein GANs. Advances in neural information processing systems 30 (2017)

    2017

  46. [54]

    ACM Computing Surveys 56(4), 1–39 (2023)

    Yang, L., Zhang, Z., Song, Y., Hong, S., Xu, R., Zhao, Y., Zhang, W., Cui, B., Yang, M.-H.: Diffusion models: A comprehensive survey of methods and applications. ACM Computing Surveys 56(4), 1–39 (2023)

    2023

  47. [55]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Shipard, J., Wiliem, A., Thanh, K.N., Xiang, W., Fookes, C.: Diversity is 24 definitely needed: Improving model-agnostic zero-shot classification via stable dif- fusion. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 769–778 (2023)

    2023

  48. [56]

    arXiv preprint arXiv:2301.04802 (2023)

    Akrout, M., Gyepesi, B., Holl´ o, P., Po´ or, A., Kincs˝ o, B., Solis, S., Cirone, K., Kawahara, J., Slade, D., Abid, L., et al.: Diffusion-based Data Augmentation for Skin Disease Classification: Impact Across Original Medical Datasets to Fully Synthetic Images. arXiv preprin...

    2023 arXiv

  49. [57]

    IEEE transactions on image processing 13(4), 600–612 (2004)

    Wang, Z., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P.: Image quality assess- ment: from error visibility to structural similarity. IEEE transactions on image processing 13(4), 600–612 (2004)

    2004

  50. [58]

    In: BIOSTEC (2), pp

    Santangelo, G., Nicora, G., Bellazzi, R., Dagliati, A., et al.: Synthcheck: A dash- board for synthetic data quality assessment. In: BIOSTEC (2), pp. 246–256 (2024)

    2024

  51. [59]

    ACM Journal of Data and Information Quality16(1), 1–33 (2024)

    Breuer, T., Fuhr, N., Schaer, P.: Validating synthetic usage data in living lab environments. ACM Journal of Data and Information Quality16(1), 1–33 (2024)

    2024

  52. [60]

    International Journal of Medical Informatics, 105413 (2024)

    Vallevik, V.B., Babic, A., Marshall, S.E., Severin, E., Brøgger, H.M., Alagarat- nam, S., Edwin, B., Veeraragavan, N.R., Befring, A.K., Nyg ˚ ard, J.F.: Can I trust my fake data–A comprehensive quality assessment framework for synthetic tab- ular data in healthcare. Internatio...

    2024

  53. [61]

    Advances in neural information processing systems 31 (2018)

    Lucic, M., Kurach, K., Michalski, M., Gelly, S., Bousquet, O.: Are GANs created equal? A large-scale study. Advances in neural information processing systems 31 (2018)

    2018

  54. [62]

    Computer vision and image understanding 179, 41–65 (2019)

    Borji, A.: Pros and cons of GAN evaluation measures. Computer vision and image understanding 179, 41–65 (2019)

    2019

  55. [63]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp

    Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., Wojna, Z.: Rethinking the incep- tion architecture for computer vision. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2818–2826 (2016)

    2016

  56. [64]

    arXiv preprint arXiv:1801.01973 (2018)

    Barratt, S., Sharma, R.: A note on the inception score. arXiv preprint arXiv:1801.01973 (2018)

    2018 arXiv

  57. [65]

    Ravuri, S., Vinyals, O.: Seeing is not necessarily believing: Limitations of BigGANs for data augmentation (2019)

    2019

  58. [66]

    In: 2015 IEEE International Conference on Image Processing (ICIP), pp

    Pambrun, J.-F., Noumeir, R.: Limitations of the ssim quality metric in the con- text of diagnostic imaging. In: 2015 IEEE International Conference on Image Processing (ICIP), pp. 2960–2963 (2015). IEEE 25

    2015

  59. [67]

    Electronics letters 44(13), 800–801 (2008)

    Huynh-Thu, Q., Ghanbari, M.: Scope of validity of PSNR in image/video quality assessment. Electronics letters 44(13), 800–801 (2008)

    2008

  60. [68]

    arXiv preprint arXiv:2308.00402 (2023)

    Scholz, D., Wiestler, B., Rueckert, D., Menten, M.J.: Metrics to Quantify Global Consistency in Synthetic Medical Images. arXiv preprint arXiv:2308.00402 (2023)

    2023 arXiv

  61. [69]

    Mahon, L., Lukasiewicz, T.: Measuring Image Complexity as a Discrete Hierarchy using MDL Clustering (2022)

    2022

  62. [70]

    The annals of mathematical statistics 22(1), 79–86 (1951)

    Kullback, S., Leibler, R.A.: On information and sufficiency. The annals of mathematical statistics 22(1), 79–86 (1951)

    1951

  63. [71]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp

    Rombach, R., Blattmann, A., Lorenz, D., Esser, P., Ommer, B.: High-Resolution Image Synthesis With Latent Diffusion Models. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 10684– 10695 (2022) 26

    2022

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.