Abstract
This paper makes a first attempt to bring the Shape from Polarization (SfP) problem to the realm of deep learning. The previous state-of-the-art methods for SfP have been purely physics-based. We see value in these principled models, and blend these physical models as priors into a neural network architecture. This proposed approach achieves results that exceed the previous state-of-the-art on a challenging dataset we introduce. This dataset consists of polarization images taken over a range of object textures, paints, and lighting conditions. We report that our proposed method achieves the lowest test error on each tested condition in our dataset, showing the value of blending data-driven and physics-driven approaches.
A. Gilbert and F. Wang—Equal contribution.
Project page: https://visual.ee.ucla.edu/deepsfp.htm.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
For a detailed discussion of other sources of noise please refer to Schechner [47].
- 2.
The dataset is available at: https://visual.ee.ucla.edu/deepsfp.htm.
- 3.
- 4.
MAE is the most commonly reported measure for surface normal reconstruction, but in many cases it is a deceptive metric. We find that a few outliers in high-frequency regions can skew the MAE for entire reconstructions. Accordingly, we emphasize the qualitative comparisons of the proposed method to its physics-based counterparts.
References
Atkinson, G.A.: Polarisation photometric stereo. Comput. Vis. Image Understand. 160, 158–167 (2017)
Atkinson, G.A., Ernst, J.D.: High-sensitivity analysis of polarization by surface reflection. Mach. Vis. Appl. 29, 1171–1189 (2018)
Atkinson, G.A., Hancock, E.R.: Multi-view surface reconstruction using polarization. In: ICCV (2005)
Atkinson, G.A., Hancock, E.R.: Recovery of surface orientation from diffuse polarization. In: IEEE TIP (2006)
Baek, S.H., Jeon, D.S., Tong, X., Kim, M.H.: Simultaneous acquisition of polarimetric SVBRDF and normals. In: ACM SIGGRAPH (TOG) (2018)
Berger, K., Voorhies, R., Matthies, L.H.: Depth from stereo polarization in specular scenes for urban robotics. In: ICRA (2017)
Chen, G., Han, K., Wong, K.-Y.K.: PS-FCN: a flexible learning framework for photometric stereo. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11213, pp. 3–19. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01240-3_1
Chen, L., Zheng, Y., Subpa-asa, A., Sato, I.: Polarimetric three-view geometry. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11220, pp. 21–37. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01270-0_2
Cui, Z., Gu, J., Shi, B., Tan, P., Kautz, J.: Polarimetric multi-view stereo. In: CVPR (2017)
Deschaintre, V., Aittala, M., Durand, F., Drettakis, G., Bousseau, A.: Single-image SVBRDF capture with a rendering-aware deep network. In: ACM SIGGRAPH (TOG) (2018)
Drbohlav, O., Sara, R.: Unambiguous determination of shape from photometric stereo with unknown light sources. In: ICCV (2001)
Ghosh, A., Chen, T., Peers, P., Wilson, C.A., Debevec, P.: Circularly polarized spherical illumination reflectometry. In: ACM SIGGRAPH (TOG) (2010)
Ghosh, A., Fyffe, G., Tunwattanapong, B., Busch, J., Yu, X., Debevec, P.: Multiview face capture using polarized spherical gradient illumination. In: ACM SIGGRAPH (TOG) (2011)
Guarnera, G.C., Peers, P., Debevec, P., Ghosh, A.: Estimating surface normals from spherical stokes reflectance fields. In: Fusiello, A., Murino, V., Cucchiara, R. (eds.) ECCV 2012. LNCS, vol. 7584, pp. 340–349. Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-33868-7_34
Huang, G., Sun, Y., Liu, Z., Sedra, D., Weinberger, K.: Deep networks with stochastic depth. CoRR (2016)
Huynh, C.P., Robles-Kelly, A., Hancock, E.R.: Shape and refractive index recovery from single-view polarisation images. In: CVPR (2010)
Huynh, C.P., Robles-Kelly, A., Hancock, E.R.: Shape and refractive index from single-view spectro-polarimetric images. IJCV 101, 64–94 (2013). https://doi.org/10.1007/s11263-012-0546-3
Ikehata, S.: CNN-PS: CNN-based photometric stereo for general non-convex surfaces. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11219, pp. 3–19. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01267-0_1
Ioffe, S., Szegedy, C.: Batch normalization: accelerating deep network training by reducing internal covariate shift. arXiv preprint arXiv:1502.03167 (2015)
Jakob, W.: Mitsuba renderer (2010). http://www.mitsuba-renderer.org
Kadambi, A., Taamazyan, V., Shi, B., Raskar, R.: Polarized 3D: high-quality depth sensing with polarization cues. In: ICCV (2015)
Kadambi, A., Taamazyan, V., Shi, B., Raskar, R.: Depth sensing using geometrically constrained polarization normals. IJCV 125, 34–51 (2017). https://doi.org/10.1007/s11263-017-1025-7
Karpatne, A., Watkins, W., Read, J., Kumar, V.: Physics-guided neural networks (PGNN): an application in lake temperature modeling. CoRR (2017)
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)
Li, X., Dong, Y., Peers, P., Tong, X.: Modeling surface appearance from a single photograph using self-augmented convolutional neural networks. In: ACM SIGGRAPH (TOG) (2017)
Li, Z., Sunkavalli, K., Chandraker, M.: Materials for masses: SVBRDF acquisition with a single mobile phone image. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11207, pp. 74–90. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01219-9_5
Li, Z., Xu, Z., Ramamoorthi, R., Sunkavalli, K., Chandraker, M.: Learning to reconstruct shape and spatially-varying reflectance from a single image. In: ACM SIGGRAPH Asia (TOG) (2018)
Lindell, D.B., O’Toole, M., Wetzstein, G.: Single-photon 3D imaging with deep sensor fusion. In: ACM SIGGRAPH (TOG) (2018)
Lucid Vision Phoenix polarization camera (2018). https://thinklucid.com/product/phoenix-5-0-mp-polarized-model/
Lyu, Y., Cui, Z., Li, S., Pollefeys, M., Shi, B.: Reflection separation using a pair of unpolarized and polarized images. In: Advances in Neural Information Processing Systems, pp. 14559–14569 (2019)
Ma, W.C., Hawkins, T., Peers, P., Chabert, C.F., Weiss, M., Debevec, P.: Rapid acquisition of specular and diffuse normal maps from polarized spherical gradient illumination. In: Eurographics Conference on Rendering Techniques (2007)
Maeda, T., Kadambi, A., Schechner, Y.Y., Raskar, R.: Dynamic heterodyne interferometry. In: ICCP (2018)
Mahmoud, A.H., El-Melegy, M.T., Farag, A.A.: Direct method for shape recovery from polarization and shading. In: ICIP (2012)
Marco, J., et al.: Deeptof: off-the-shelf real-time correction of multipath interference in time-of-flight imaging. In: ACM SIGGRAPH (TOG) (2017)
Miyazaki, D., Kagesawa, M., Ikeuchi, K.: Transparent surface modeling from a pair of polarization images. In: PAMI (2004)
Miyazaki, D., Shigetomi, T., Baba, M., Furukawa, R., Hiura, S., Asada, N.: Surface normal estimation of black specular objects from multiview polarization images. Int. Soc. Opt. Photon. Opt. Eng. (2016)
Miyazaki, D., Tan, R.T., Hara, K., Ikeuchi, K.: Polarization-based inverse rendering from a single view. In: ICCV (2003)
Mo, Z., Shi, B., Lu, F., Yeung, S.K., Matsushita, Y.: Uncalibrated photometric stereo under natural illumination. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2936–2945 (2018)
Ngo, T.T., Nagahara, H., Taniguchi, R.: Shape and light directions from shading and polarization. In: CVPR (2015)
Park, T., Liu, M.Y., Wang, T.C., Zhu, J.Y.: Semantic image synthesis with spatially-adaptive normalization. In: CVPR (2019)
Paszke, A., et al.: Automatic differentiation in pytorch. In: NIPS-W (2017)
PolarM polarization camera (2017). http://www.4dtechnology.com/products/polarimeters/polarcam/
Riviere, J., Reshetouski, I., Filipi, L., Ghosh, A.: Polarization imaging reflectometry in the wild. In: ACM SIGGRAPH (TOG) (2017)
Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-24574-4_28
Santo, H., Samejima, M., Sugano, Y., Shi, B., Matsushita, Y.: Deep photometric stereo network. In: ICCV Workshops (2017)
Satat, G., Tancik, M., Gupta, O., Heshmat, B., Raskar, R.: Object classification through scattering media with deep learning on time resolved measurement. OSA Opt. Exp. 25, 17466–17479 (2017)
Schechner, Y.Y.: Self-calibrating imaging polarimetry. In: ICCP (2015)
Sengupta, S., Kanazawa, A., Castillo, C.D., Jacobs, D.W.: SfSnet: learning shape, reflectance and illuminance of faces in the wild. In: CVPR (2018)
Shi, B., Mo, Z., Wu, Z., Duan, D., Yeung, S.K., Tan, P.: A benchmark dataset and evaluation for non-Lambertian and uncalibrated photometric stereo. In: PAMI (2019)
SHINING 3D scanner (2018). https://www.einscan.com/einscan-se-sp
Smith, W.A.P., Ramamoorthi, R., Tozza, S.: Linear depth estimation from an uncalibrated, monocular polarisation image. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9912, pp. 109–125. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46484-8_7
Smith, W.A.P., Ramamoorthi, R., Tozza, S.: Height-from-polarisation with unknown lighting or albedo. In: PAMI (2018)
Srivastava, R.K., Greff, K., Schmidhuber, J.: Highway networks. CoRR (2015)
Su, S., Heide, F., Wetzstein, G., Heidrich, W.: Deep end-to-end time-of-flight imaging. In: CVPR (2018)
Tancik, M., Satat, G., Raskar, R.: Flash photography for data-driven hidden scene recovery. arXiv preprint arXiv:1810.11710 (2018)
Tancik, M., Swedish, T., Satat, G., Raskar, R.: Data-driven non-line-of-sight imaging with a traditional camera. In: OSA Imaging and Applied Optics (2018)
Taniai, T., Maehara, T.: Neural inverse rendering for general reflectance photometric stereo. In: ICML (2018)
Teo, D., Shi, B., Zheng, Y., Yeung, S.K.: Self-calibrating polarising radiometric calibration. In: CVPR (2018)
Tozza, S., Smith, W.A.P., Zhu, D., Ramamoorthi, R., Hancock, E.R.: Linear differential constraints for photo-polarimetric height estimation. In: ICCV (2017)
Wolff, L.B.: Polarization vision: a new sensory approach to image understanding. Image Vis. Comput. 15, 81–93 (1997)
Xiong, Y., Chakrabarti, A., Basri, R., Gortler, S.J., Jacobs, D.W., Zickler, T.: From shading to local shape. IEEE Trans. Pattern Anal. Mach. Intell. 37(1), 67–79 (2014)
Yang, L., Tan, F., Li, A., Cui, Z., Furukawa, Y., Tan, P.: Polarimetric dense monocular SLAM. In: CVPR (2018)
Ye, W., Li, X., Dong, Y., Peers, P., Tong, X.: Single image surface appearance modeling with self-augmented CNNs and inexact supervision. Wiley Online Library Computer Graphics Forum (2018)
Zhu, D., Smith, W.A.P.: Depth from a polarisation + RGB stereo pair. In: CVPR (2019)
Acknowledgements
The work of UCLA authors was supported by a Sony Imaging Young Faculty Award, Google Faculty Award, and the NSF CRII Research Initiation Award (IIS 1849941). The work of Peking University authors was supported by National Natural Science Foundation of China (61872012, 41842048, 41571432), National Key R&D Program of China (2019YFF0302902, 2017YFB0503004), Beijing Academy of Artificial Intelligence (BAAI), and Education Department Project of Guizhou Province.
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Ba, Y. et al. (2020). Deep Shape from Polarization. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, JM. (eds) Computer Vision – ECCV 2020. ECCV 2020. Lecture Notes in Computer Science(), vol 12369. Springer, Cham. https://doi.org/10.1007/978-3-030-58586-0_33
Download citation
DOI: https://doi.org/10.1007/978-3-030-58586-0_33
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-58585-3
Online ISBN: 978-3-030-58586-0
eBook Packages: Computer ScienceComputer Science (R0)