DeceptionNet: Network-Driven Domain Randomization
Open accessThe DOI matches its Crossref record; 23 of the 23 references it cites were found in a registry or an index.
Abstract
We present a novel approach to tackle domain adaptation between synthetic and real data. Instead, of employing "blind" domain randomization, i.e., augmenting synthetic renderings with random backgrounds or changing illumination and colorization, we leverage the task network as its own adversarial guide toward useful augmentations that maximize the uncertainty of the output. To this end, we design a min-max optimization scheme where a given task competes against a special deception network to minimize the task error subject to the specific constraints enforced by the deceiver. The deception network samples from a family of differentiable pixel-level perturbations and exploits the task architecture to find the most destructive augmentations. Unlike GAN-based approaches that require unlabeled data from the target domain, our method achieves robust mappings that scale well to multiple target distributions from source data alone. We apply our framework to the tasks of digit recognition on enhanced MNIST variants, classification and object pose estimation on the Cropped LineMOD dataset as well as semantic segmentation on the Cityscapes dataset and compare it to a number of domain adaptation approaches, thereby demonstrating similar results with superior generalization capabilities.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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
Domain-Adversarial Training of Neural Networks
Ganin, Y., Ustinova, E., Ajakan, H., Germain, P., Larochelle, H., Laviolette, F., Marchand, M., & Lempitsky, V. (2017) · Advances in Computer Vision and Pattern Recognition · doi:10.1007/978-3-319-58347-1_10 · Cited by 8,318
Image Style Transfer Using Convolutional Neural Networks
Gatys, L. A., Ecker, A. S., & Bethge, M. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.265 · Cited by 6,025
Adversarial Discriminative Domain Adaptation
Tzeng, E., Hoffman, J., Saenko, K., & Darrell, T. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.316 · Cited by 5,154 · cited 3×
“Recent unsupervised approaches are mostly based on generalized adversarial networks (GANs) [4, 22, 26, 46, 49, 24, 43, 36, 1] and although these methods perform proper target domain transfers, they can overfit to the chosen target domain and exhibit a decline in performance for unfamiliar out-of-distribution samples.”
BackgroundMethodJaderberg, M., Simonyan, K., Zisserman, A., & Kavukcuoglu, K. (2015) · arXiv · doi:10.48550/arxiv.1506.02025 · Cited by 5,137
Learning Transferable Features with Deep Adaptation Networks
Long, M., Cao, Y., Wang, J., & Jordan, M. I. (2015) · arXiv · doi:10.48550/arxiv.1502.02791 · Cited by 4,505
Learning and Transferring Mid-level Image Representations Using Convolutional Neural Networks
Oquab, M., Bottou, L., Laptev, I., & Sivic, J. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.222 · Cited by 3,150 · cited 1×
“In the case of supervised domain adaptation approaches [54, 32, 3, 30], a certain amount of labeled data from the target domain exists, while in unsupervised approaches [13, 46, 41, 4] the target data are available but unlabeled.”
BackgroundDomain randomization for transferring deep neural networks from simulation to the real world
Tobin, J., Fong, R., Ray, A., Schneider, J., Zaremba, W., & Abbeel, P. (2017) · 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2017.8202133 · Cited by 3,124 · cited 3×
“Domain randomization methods [47, 21, 29, 52, 45] have no access to any target domain and employ the rather simple technique of randomly perturbing (synthetic) source data during training to make the tasks networks robust to perceptual differences.”
BackgroundMethodOne Pixel Attack for Fooling Deep Neural Networks
Su, J., Vargas, D. V., & Sakurai, K. (2019) · IEEE Transactions on Evolutionary Computation · doi:10.1109/tevc.2019.2890858 · Cited by 2,406 · cited 1×
“Last but not least, results from pixellevel adversarial attacks [6, 44] suggest the existence of architecture-dependent effects that cannot be addressed by ”blind” domain randomization for robust transfer.”
BackgroundRos, G., Sellart, L., Materzynska, J., Vazquez, D., & Lopez, A. M. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.352 · Cited by 2,338
Deep domain confusion Maximizing for domain invariance
Tzeng, E., Hoffman, J., Zhang, N., Saenko, K., & Darrell, T. (2014) · arXiv · doi:10.48550/arxiv.1412.3474 · Cited by 2,285
Research works printed but not linked: 36
- sadeghi, 2016, Cad2rl: Real single-image flight without a single real image, arXiv preprint arXiv 1611 04201
- 10.1145/360825.360839 · DOI · Paper page
- 10.1109/ICCV.2017.609 · DOI · Paper page
- ratner, 2017, Learning to compose domain-specific transformations for data augmentation, Advances in neural information processing systems
- 10.1109/CVPR.2018.00490 · DOI
- 10.1109/IROS40897.2019.8967829 · DOI
- mahler, 2017, Learning deep policies for robot bin picking by simulating robust grasping sequences, CoRL
- 10.1007/978-3-030-01264-9_49 · DOI
- 10.1109/TPAMI.2010.161 · DOI · Paper page
- antoniou, 2017, Data augmentation generative adversarial networks, arXiv preprint arXiv 1711 04340
- ledig, 2016, Photo-realistic single image super-resolution using a generative adversarial network, CoRR
- lee, 2019, Spigan: Privileged adversarial learning from simulation, ICLRE
- liu, 2016, Coupled generative adversarial networks, CoRR
- lin, 2014, Microsoft coco: common objects in context, ECCV
- 10.1109/CVPR.2015.7298930 · DOI · Paper page
- 10.1109/ICCV.2017.244 · DOI · Paper page
- 10.1109/ICCV.2017.223 · DOI · Paper page
- 10.1109/ICCV.2019.00203 · DOI · Paper page
- 10.1109/3DV.2018.00012 · DOI
- 10.1109/CVPR.2016.350 · DOI · Paper page
- devries, 2017, Dataset augmentation in feature space, arXiv preprint arXiv 1702 05464
- 10.1109/CVPR.2018.00395 · DOI · Paper page
- ganin, 2015, Unsupervised domain adaptation by backpropagation, ICML
- 10.1109/WACV.2017.103 · DOI · Paper page
- hoffman, 2016, Fcns in the wild: Pixel-level adversarial and constraint-based adaptation, arXiv preprint arXiv 1612 02649
- 10.1007/s10851-009-0161-2 · DOI · Paper page
- babenko, 2014, Neural codes for image retrieval, CoRR
- bousmalis, 2016, Domain separation networks, NIPS
- 10.1109/CVPR.2018.00823 · DOI · Paper page
- 10.1109/CVPR.2019.00194 · DOI · Paper page
- 10.1109/ICCV.2017.220 · DOI · Paper page
- taigman, 2017, Unsupervised cross-domain image generation, ICLRE
- 10.1109/CVPR.2018.00780 · DOI · Paper page
- 10.1109/ICDAR.2003.1227801 · DOI · Paper page
- 10.1109/CVPR.2017.241 · DOI · Paper page
- 10.3389/frobt.2018.00066 · DOI · Paper page
Cited by
Solving Rubik's Cube with a Robot Hand
OpenAI, Akkaya, I., Andrychowicz, M., Chociej, M., Litwin, M., McGrew, B., Petron, A., Paino, A., Plappert, M., Powell, G., et al. (2019) · arXiv · doi:10.48550/arxiv.1910.07113 · Cited by 602
Liu, Q., Chen, C., Qin, J., Dou, Q., & Heng, P.-A. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.00107 · Cited by 541
Synthetic Data for Deep Learning
Nikolenko, S. I. (2021) · Springer Optimization and Its Applications · doi:10.1007/978-3-030-75178-4 · Cited by 423
Learning to Generate Novel Domains for Domain Generalization
Zhou, K., Yang, Y., Hospedales, T., & Xiang, T. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58517-4_33 · Cited by 382
A Simple Semi-Supervised Learning Framework for Object Detection
Sohn, K., Zhang, Z., Li, C., Zhang, H., Lee, C., & Pfister, T. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2005.04757 · Cited by 306
SPEED+: Next-Generation Dataset for Spacecraft Pose Estimation across Domain Gap
Park, T. H., Martens, M., Lecuyer, G., Izzo, D., & D'Amico, S. (2022) · 2022 IEEE Aerospace Conference (AERO) · doi:10.1109/aero53065.2022.9843439 · Cited by 170
Occlusion-Aware Self-Supervised Monocular 6D Object Pose Estimation
Wang, G., Manhardt, F., Liu, X., Ji, X., & Tombari, F. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3136301 · Cited by 69
Metadata
- Authors
- Zakharov, S., Kehl, W., & Ilic, S.
- Venue
- 2019 IEEE/CVF International Conference on Computer Vision (ICCV)
- Publisher
- IEEE
- Published
- 2019-10
- Pages
- 532-541
- OpenAlex
- W2926800893
- Cited by
- 78
- References
- 23
- FWCI
- 4.91
- Type
- conference paper
- Language
- en
- Open access
- green
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.