Abstract
A critical limitation of current methods based on Neural Radiance Fields (NeRF) is that they are unable to quantify the uncertainty associated with the learned appearance and geometry of the scene. This information is paramount in real applications such as medical diagnosis or autonomous driving where, to reduce potentially catastrophic failures, the confidence on the model outputs must be included into the decision-making process. In this context, we introduce Conditional-Flow NeRF (CF-NeRF), a novel probabilistic framework to incorporate uncertainty quantification into NeRF-based approaches. For this purpose, our method learns a distribution over all possible radiance fields modelling the scene which is used to quantify the uncertainty associated with the modelled scene. In contrast to previous approaches enforcing strong constraints over the radiance field distribution, CF-NeRF learns it in a flexible and fully data-driven manner by coupling Latent Variable Modelling and Conditional Normalizing Flows. This strategy allows to obtain reliable uncertainty estimation while preserving model expressivity. Compared to previous state-of-the-art methods proposed for uncertainty quantification in NeRF, our experiments show that the proposed method achieves significantly lower prediction errors and more reliable uncertainty values for synthetic novel view and depth-map estimation.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
References
Aralikatti, R., Margam, D., Sharma, T., Thanda, A., Venkatesan, S.: Global SNR estimation of speech signals using entropy and uncertainty estimates from dropout networks. In: INTERSPEECH (2018)
Ashukha, A., Lyzhov, A., Molchanov, D., Vetrov, D.P.: Pitfalls of in-domain uncertainty estimation and ensembling in deep learning. In: ICLR (2020)
Bae, G., Budvytis, I., Cipolla, R.: Estimating and exploiting the aleatoric uncertainty in surface normal estimation. In: ICCV (2021)
van den Berg, R., Hasenclever, L., Tomczak, J., Welling, M.: Sylvester normalizing flows for variational inference. In: Proceedings of the Conference on Uncertainty in Artificial Intelligence (UAI) (2018)
Bi, S., et al.: Deep reflectance volumes: relightable reconstructions from multi-view photometric images. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12348, pp. 294–311. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58580-8_18
Blundell, C., Cornebise, J., Kavukcuoglu, K., Wierstra, D.: Weight uncertainty in neural networks. In: ICML (2015)
Brachmann, E., Michel, F., Krull, A., Yang, M.Y., Gumhold, S., Rother, C.: Uncertainty-driven 6D pose estimation of objects and scenes from a single RGB image. In: CVPR (2016)
Djuric, N., et al.: Uncertainty-aware short-term motion prediction of traffic actors for autonomous driving. In: WACV (2020)
Eigen, D., Puhrsch, C., Fergus, R.: Depth map prediction from a single image using a multi-scale deep network. In: NIPS (2014)
Gafni, G., Thies, J., Zollhöfer, M., Nießner, M.: Dynamic neural radiance fields for monocular 4D facial avatar reconstruction. In: CVPR (2021)
Gal, Y., Ghahramani, Z.: Dropout as a Bayesian approximation: representing model uncertainty in deep learning. In: ICML (2016)
Geifman, Y., Uziel, G., El-Yaniv, R.: Bias-reduced uncertainty estimation for deep neural classifiers. In: ICLR (2019)
Hendrycks, D., Mazeika, M., Kadavath, S., Song, D.X.: Using self-supervised learning can improve model robustness and uncertainty. In: NeurIPS (2019)
Hernández, S., Vergara, D., Valdenegro-Toro, M., Jorquera, F.: Improving predictive uncertainty estimation using Dropout-Hamiltonian Monte Carlo. Soft. Comput. 24, 4307–4322 (2020)
Jiakai, Z., et al.: Editable free-viewpoint video using a layered neural representation. In: ACM SIGGRAPH (2021)
Kajiya, J., von Herzen, B.: Ray tracing volume densities. ACM SIGGRAPH Comput. Graph. 18, 165–174 (1984)
Kirsch, W.: An elementary proof of de Finetti’s theorem. Stat. Probab. Lett. 151(C), 84–88 (2019)
Klokov, R., Boyer, E., Verbeek, J.: Discrete point flow networks for efficient point cloud generation. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12368, pp. 694–710. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58592-1_41
Kosiorek, A.R., et al.: NERF-VAE: a geometry aware 3D scene generative model. In: ICML (2021)
Lakshminarayanan, B., Pritzel, A., Blundell, C.: Simple and scalable predictive uncertainty estimation using deep ensembles. In: Advances in Neural Information Processing Systems (2017)
Li, Y., Li, S., Sitzmann, V., Agrawal, P., Torralba, A.: 3D neural scene representations for visuomotor control. In: CoRL (2021)
Li, Z., Niklaus, S., Snavely, N., Wang, O.: Neural scene flow fields for space-time view synthesis of dynamic scenes. In: CVPR (2021)
Liu, J., Paisley, J.W., Kioumourtzoglou, M.A., Coull, B.: Accurate uncertainty estimation and decomposition in ensemble learning. In: NeurIPS (2019)
Liu, L., Gu, J., Lin, K.Z., Chua, T.S., Theobalt, C.: Neural sparse voxel fields. In: NeurIPS (2020)
Liu, S., Zhang, X., Zhang, Z., Zhang, R., Zhu, J.Y., Russell, B.: Editing conditional radiance fields. In: ICCV (2021)
Lombardi, S., Simon, T., Saragih, J., Schwartz, G., Lehrmann, A., Sheikh, Y.: Neural volumes: learning dynamic renderable volumes from images. ACM Trans. Graph. 38(4), 65:1–65:14 (2019)
Long, K., Qian, C., Cortes, J., Atanasov, N.: Learning barrier functions with memory for robust safe navigation. IEEE Robot. Autom. Lett. 6(3), 1 (2021)
Loquercio, A., Segu, M., Scaramuzza, D.: A general framework for uncertainty estimation in deep learning. IEEE Robot. Autom. Lett. 5(2), 3153–3160 (2020)
Malinin, A., Gales, M.J.F.: Predictive uncertainty estimation via prior networks. In: NeurIPS (2018)
Martin-Brualla, R., Radwan, N., Sajjadi, M.S.M., Barron, J.T., Dosovitskiy, A., Duckworth, D.: NeRF in the wild: neural radiance fields for unconstrained photo collections. In: CVPR (2021)
Mescheder, L., Oechsle, M., Niemeyer, M., Nowozin, S., Geiger, A.: Occupancy networks: learning 3D reconstruction in function space. In: CVPR (2019)
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
Neal, R.M.: Bayesian Learning for Neural Networks. Springer, New York (1995)
Neapolitan, R.E.: Learning Bayesian networks. In: Proceedings of the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD) (2007)
Neff, T., et al.: DONeRF: towards real-time rendering of compact neural radiance fields using depth oracle networks. Comput. Graph. Forum 40(4), 45–59 (2021)
Niemeyer, M., Barron, J.T., Mildenhall, B., Sajjadi, M.S.M., Geiger, A., Radwan, N.: RegNeRF: regularizing neural radiance fields for view synthesis from sparse inputs. arXiv abs/2112.00724 (2021)
Niemeyer, M., Mescheder, L., Oechsle, M., Geiger, A.: Occupancy flow: 4D reconstruction by learning particle dynamics. In: ICCV (2019)
Park, J.J., Florence, P., Straub, J., Newcombe, R.A., Lovegrove, S.: DeepSDF: learning continuous signed distance functions for shape representation. In: CVPR (2019)
Park, K., et al.: Nerfies: deformable neural radiance fields. In: ICCV (2021)
Peng, L.W., Shamsuddin, S.M.: 3D object reconstruction and representation using neural networks. In: Proceedings of the 2nd International Conference on Computer Graphics and Interactive Techniques in Australasia and South East Asia (2004)
Peng, S., et al.: Animatable neural radiance fields for human body modeling. In: CVPR (2021)
Poggi, M., Aleotti, F., Tosi, F., Mattoccia, S.: On the uncertainty of self-supervised monocular depth estimation. In: CVPR (2020)
Pumarola, A., Corona, E., Pons-Moll, G., Moreno-Noguer, F.: D-NeRF: neural radiance fields for dynamic scenes. In: CVPR (2021)
Pumarola, A., Popov, S., Moreno-Noguer, F., Ferrari, V.: C-flow: conditional generative flow models for images and 3D point clouds. In: CVPR (2020)
Qu, C., Liu, W., Taylor, C.J.: Bayesian deep basis fitting for depth completion with uncertainty. In: ICCV (2021)
Raj, A., et al.: PVA: pixel-aligned volumetric avatars. In: CVPR (2021)
Rebain, D., Jiang, W., Yazdani, S., Li, K., Yi, K.M., Tagliasacchi, A.: DeRF: decomposed radiance fields. In: CVPR (2020)
Ritter, H., Botev, A., Barber, D.: A scalable laplace approximation for neural networks. In: ICLR (2018)
Saito, S., Yang, J., Ma, Q., Black, M.J.: SCANimate: weakly supervised learning of skinned clothed avatar networks. In: CVPR (2021)
Schwarz, K., Liao, Y., Niemeyer, M., Geiger, A.: GRAF: generative radiance fields for 3D-aware image synthesis. In: NIPS (2020)
Shamsi, A., et al.: An uncertainty-aware transfer learning-based framework for Covid-19 diagnosis. IEEE Trans. Neural Netw. Learn. Syst. 32, 1408–1417 (2021)
Shen, J., Ruiz, A., Agudo, A., Moreno-Noguer, F.: Stochastic neural radiance fields: quantifying uncertainty in implicit 3D representations. In: 3DV (2021)
Sitzmann, V., Martel, J.N.P., Bergman, A.W., Lindell, D.B., Wetzstein, G.: Implicit neural representations with periodic activation functions. In: NIPS (2020)
Sitzmann, V., Zollhöfer, M., Wetzstein, G.: Scene representation networks: continuous 3D-structure-aware neural scene representations. In: NIPS (2019)
Su, S.Y., Yu, F., Zollhöfer, M., Rhodin, H.: A-NeRF: articulated neural radiance fields for learning human shape, appearance, and pose. In: NIPS (2021)
Tancik, M., et al.: Learned initializations for optimizing coordinate-based neural representations. In: CVPR (2021)
Tancik, M., et al.: Fourier features let networks learn high frequency functions in low dimensional domains. In: NIPS (2020)
Teye, M., Azizpour, H., Smith, K.: Bayesian uncertainty estimation for batch normalized deep networks. In: ICML (2018)
Wang, Z., Simoncelli, E., Bovik, A.: Multiscale structural similarity for image quality assessment. In: The Thirty-Seventh Asilomar Conference on Signals, Systems Computers, vol. 2, pp. 1398–1402 (2003)
Wei, Y., Liu, S., Rao, Y., Zhao, W., Lu, J., Zhou, J.: NerfingMVS: guided optimization of neural radiance fields for indoor multi-view stereo. In: ICCV (2021)
Xie, Y., et al.: Neural fields in visual computing and beyond (2021)
Xu, H., et al.: Digging into uncertainty in self-supervised multi-view stereo. In: ICCV (2021)
Yang, G., Huang, X., Hao, Z., Liu, M.Y., Belongie, S.J., Hariharan, B.: PointFlow: 3D point cloud generation with continuous normalizing flows. In: ICCV (2019)
Yen-Chen, L., Florence, P., Barron, J.T., Rodriguez, A., Isola, P., Lin, T.Y.: iNeRF: inverting neural radiance fields for pose estimation. In: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (2021)
Yu, A., Ye, V., Tancik, M., Kanazawa, A.: pixelNeRF: neural radiance fields from one or few images. In: CVPR (2021)
Yücer, K., Sorkine-Hornung, A., Wang, O., Sorkine-Hornung, O.: Efficient 3D object segmentation from densely sampled light fields with applications to 3D reconstruction. ACM Trans. Graph. 35(3) (2016)
Zhang, K., Riegler, G., Snavely, N., Koltun, V.: NeRF++: analyzing and improving neural radiance fields. arXiv abs/2010.07492 (2020)
Zhang, R., Isola, P., Efros, A.A., Shechtman, E., Wang, O.: The unreasonable effectiveness of deep features as a perceptual metric. In: CVPR (2018)
Zhu, L., et al.: RGB-D local implicit function for depth completion of transparent objects. In: CVPR (2021)
Acknowledgements
This work is supported partly by the Chinese Scholarship Council (CSC) under grant (201906120031), by the Spanish government under project MoHuCo PID2020-120049RB-I00 and the Chistera project IPALM PCI2019-103386. We also thank Nvidia for hardware donation under the GPU grant program.
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.
Supplementary material 2 (mp4 1120 KB)
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Shen, J., Agudo, A., Moreno-Noguer, F., Ruiz, A. (2022). Conditional-Flow NeRF: Accurate 3D Modelling with Reliable Uncertainty Quantification. 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 13663. Springer, Cham. https://doi.org/10.1007/978-3-031-20062-5_31
Download citation
DOI: https://doi.org/10.1007/978-3-031-20062-5_31
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-20061-8
Online ISBN: 978-3-031-20062-5
eBook Packages: Computer ScienceComputer Science (R0)