Abstract
Training computer vision models usually requires collecting and labeling vast amounts of imagery under a diverse set of scene configurations and properties. This process is incredibly time-consuming, and it is challenging to ensure that the captured data distribution maps well to the target domain of an application scenario. Recently, synthetic data has emerged as a way to address both of these issues. However, existing approaches either require human experts to manually tune each scene property or use automatic methods that provide little to no control; this requires rendering large amounts of random data variations, which is slow and is often suboptimal for the target domain. We present the first fully differentiable synthetic data pipeline that uses Neural Radiance Fields (NeRFs) in a closed-loop with a target application’s loss function. Our approach generates data on-demand, with no human labor, to maximize accuracy for a target task. We illustrate the effectiveness of our method on synthetic and real-world object detection tasks. We also introduce a new “YCB-in-the-Wild” dataset and benchmark that provides a test scenario for object detection with varied poses in real-world environments. Code and data could be found at
.
H. Behl and J. Xu—Equal contribution as second author.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
For simplicity, we have dropped the dependence of loss \(\ell \) on labels y.
References
Jahanian, A., Lucy Chai, P.I.: On the "steerability" of generative adversarial networks. CoRR (2019)
Barbu, A., et al.: Objectnet: A large-scale bias-controlled dataset for pushing the limits of object recognition models. In: Wallach, H., Larochelle, H., Beygelzimer, A., d’Alché-Buc, F., Fox, E., Garnett, R. (eds.) Advances in Neural Information Processing Systems. vol. 32. Curran Associates, Inc. (2019), https://proceedings.neurips.cc/paper/2019/file/97af07a14cacba681feacf3012730892-Paper.pdf
Barbu, A., et al.: Objectnet: A large-scale bias-controlled dataset for pushing the limits of object recognition models. In: Advances in Neural Information Processing Systems, vol. 32 (2019)
Behl, H.S., Baydin, A.G., Gal, R., Torr, P.H.S., Vineet, V.: Autosimulate: (quickly) learning synthetic data generation. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12367, pp. 255–271. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58542-6_16
Bi, S., et al.: Neural reflectance fields for appearance acquisition. arXiv preprint arXiv:2008.03824 (2020)
Brock, A., Donahue, J., Simonyan, K.: Large scale GAN training for high fidelity natural image synthesis. In: International Conference on Learning Representations (2019). https://openreview.net/forum?id=B1xsqj09Fm
Calli, B., Walsman, A., Singh, A., Srinivasa, S., Abbeel, P., Dollar, A.M.: Benchmarking in manipulation research: The ycb object and model set and benchmarking protocols. arXiv preprint arXiv:1502.03143 (2015)
Colson, B., Marcotte, P., Savard, G.: An overview of bilevel optimization. Ann. Oper. Res. 153(1), 235–256 (2007)
Danilo Jimenez Rezende, S.M.: Variational inference with normalizing flows. In: ICML (2015)
Denninger, M., et al.: Blenderproc. arXiv preprint arXiv:1911.01911 (2019)
Devaranjan, J., Kar, A., Fidler, S.: Meta-Sim2: Unsupervised learning of scene structure for synthetic data generation. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12362, pp. 715–733. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58520-4_42
Diederik Kingma, M.W.: Autoencoding variational bayes. In: ICLR (2014)
Doersch, C., Zisserman, A.: Sim2real transfer learning for 3d human pose estimation: motion to the rescue. In: NeurIPS (2019)
Dwibedi, D., Misra, I., Hebert, M.: Cut, paste and learn: Surprisingly easy synthesis for instance detection. In: ICCV (2017)
Franceschi, L., Frasconi, P., Salzo, S., Grazzi, R., Pontil, M.: Bilevel programming for hyperparameter optimization and meta-learning. In: International Conference on Machine Learning, pp. 1568–1577. PMLR (2018)
Gafni, G., Thies, J., Zollhofer, M., Nießner, M.: Dynamic neural radiance fields for monocular 4d facial avatar reconstruction. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 8649–8658 (2021)
Ganin, Y., Kulkarni, T., Babuschkin, I., Eslami, S.M.A., Vinyals, O.: Synthesizing programs for images using reinforced adversarial learning. In: ICML (2018)
Ge, Y., Abu-El-Haija, S., Xin, G., Itti, L.: Zero-shot synthesis with group-supervised learning. arXiv preprint arXiv:2009.06586 (2020)
Ge, Y., Xu, J., Zhao, B.N., Itti, L., Vineet, V.: Dall-e for detection: Language-driven context image synthesis for object detection. arXiv preprint arXiv:2206.09592 (2022)
Ge, Y., Zhao, J., Itti, L.: Pose augmentation: Class-agnostic object pose transformation for object recognition. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12373, pp. 138–155. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58604-1_9
Georgiev, I., et al.: Arnold: A brute-force production path tracer. TOG 37, 1–12 (2018)
Goodfellow, I., et al.: Generative adversarial nets. In: Advances in neural information processing systems, pp. 2672–2680 (2014)
Handa, A., Patraucean, V., Badrinarayanan, V., Stent, S., Cipolla, R.: Understanding real world indoor scenes with synthetic data. In: CVPR (2016)
He, K., Gkioxari, G., Dollár, P., Girshick, R.: Mask r-cnn. In: ICCV (2017)
Higgins, I., et al.: beta-vae: Learning basic visual concepts with a constrained variational framework. In: 5th International Conference on Learning Representations, ICLR 2017, Toulon, France (2017)
Hodaň, T., et al.: BOP: Benchmark for 6d object pose estimation. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11214, pp. 19–35. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01249-6_2
Hodaň, T., et al.: Photorealistic image synthesis for object instance detection. In: ICIP (2019)
Ilg, E., Mayer, N., Saikia, T., Keuper, M., Dosovitskiy, A., Brox, T.: Flownet 2.0: Evolution of optical flow estimation with deep networks. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, Honolulu, HI, USA, 21–26 July 2017, pp. 1647–1655 (2017). https://doi.org/10.1109/CVPR.2017.179
Jang, W., Agapito, L.: Codenerf: Disentangled neural radiance fields for object categories. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 12949–12958 (2021)
Kar, A., et al.: Meta-sim: Learning to generate synthetic datasets. In: ICCV (2019)
Lin, T.Y., Goyal, P., Girshick, R., He, K., Dollár, P.: Focal loss for dense object detection. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 2980–2988 (2017)
Louppe, G., Cranmer, K.: Adversarial variational optimization of non-differentiable simulators. In: AISTATS (2019)
Martin-Brualla, R., Radwan, N., Sajjadi, M.S., Barron, J.T., Dosovitskiy, A., Duckworth, D.: Nerf in the wild: Neural radiance fields for unconstrained photo collections. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 7210–7219 (2021)
Mildenhall, B., Srinivasan, P.P., Tancik, M., Barron, J.T., Ramamoorthi, R., Ng, R.: NeRF: Representing scenes as neural radiance fields for view synthesis. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12346, pp. 405–421. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58452-8_24
Ng, A.: Mlops: From model-centric to data-centric ai. https://www.deeplearning.ai/wp-content/uploads/2021/06/MLOps-From-Model-centric-to-Data-centric-AI.pdf
Park, K., et al.: Nerfies: Deformable neural radiance fields. In: ICCV (2021)
Ren, S., He, K., Girshick, R., Sun, J.: Faster r-cnn: towards real-time object detection with region proposal networks. In: PAMI (2017)
Richter, S.R., Hayder, Z., Koltun, V.: Playing for benchmarks. In: ICCV (2017)
Richter, S.R., Vineet, V., Roth, S., Koltun, V.: Playing for data: Ground truth from computer games. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9906, pp. 102–118. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46475-6_7
Ros, G., Sellart, L., Materzynska, J., Vázquez, D., López, A.M.: The SYNTHIA dataset: A large collection of synthetic images for semantic segmentation of urban scenes. In: CVPR (2016)
Ruiz, N., Schulter, S., Chandraker, M.: Learning to simulate. In: ICLR (2019)
Shelhamer, E., Long, J., Darrell, T.: Fully convolutional networks for semantic segmentation. In: PAMI (2017)
Srinivasan, P.P., Deng, B., Zhang, X., Tancik, M., Mildenhall, B., Barron, J.T.: Nerv: Neural reflectance and visibility fields for relighting and view synthesis. In: CVPR (2021)
Tremblay, J., To, T., Birchfield, S.: Falling things: A synthetic dataset for 3d object detection and pose estimation. In: CVPR (2018)
Williams, R.J.: Simple statistical gradient-following algorithms for connectionist reinforcement learning. Mach. Learn. 8, 229–256 (1992)
Xiang, Y., Schmidt, T., Narayanan, V., Fox, D.: Posecnn: A convolutional neural network for 6d object pose estimation in cluttered scenes. arXiv preprint arXiv:1711.00199 (2017)
Xiaogang, X.u., Ying-Cong Chen, J.J.: View independent generative adversarial network for novel view synthesis. In: ICCV (2019)
Yang, D., Deng, J.: Learning to generate synthetic 3d training data through hybrid gradient. In: CVPR (2020)
Yen-Chen, L.: Nerf-pytorch. https://github.com/yenchenlin/nerf-pytorch/ (2020)
Zhang, X., Srinivasan, P.P., Deng, B., Debevec, P., Freeman, W.T., Barron, J.T.: Nerfactor: Neural factorization of shape and reflectance under an unknown illumination. ACM Trans. Graph. (TOG) 40(6), 1–18 (2021)
Zhang, Y., et al.: Physically-based rendering for indoor scene understanding using convolutional neural networks. In: CVPR (2017)
Acknowledgments
We thank Yen-Chen Lin for help on using the nerf-pytorch code. This work was supported in part by C-BRIC (one of six centers in JUMP, a Semiconductor Research Corporation (SRC) program sponsored by DARPA), DARPA (HR00112190134) and the Army Research Office (W911NF2020053). The authors affirm that the views expressed herein are solely their own, and do not represent the views of the United States government or any agency thereof.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Ge, Y. et al. (2022). Neural-Sim: Learning to Generate Training Data with NeRF. In: Avidan, S., Brostow, G., Cissé, M., Farinella, G.M., Hassner, T. (eds) Computer Vision – ECCV 2022. ECCV 2022. Lecture Notes in Computer Science, vol 13683. Springer, Cham. https://doi.org/10.1007/978-3-031-20050-2_28
Download citation
DOI: https://doi.org/10.1007/978-3-031-20050-2_28
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-20049-6
Online ISBN: 978-3-031-20050-2
eBook Packages: Computer ScienceComputer Science (R0)