Abstract
Light-field cameras capture sub-views from multiple perspectives simultaneously, with possibly reflectance variations that can be used to augment material recognition in remote sensing, autonomous driving, etc. Existing approaches for light-field based material recognition suffer from the entanglement between angular and spatial domains, leading to inefficient training which in turn limits their performances. In this paper, we propose an approach that achieves decoupling of angular and spatial information by establishing correspondences in the angular domain, then employs regularization to enforce a rotational invariance. As opposed to relying on the Lambertian surface assumption, we align the angular domain by estimating sub-pixel displacements using the Fourier transform. The network takes sparse inputs, i.e. sub-views along particular directions, to gain structural information about the angular domain. A novel regularization technique further improves generalization by weight sharing and max-pooling among different directions. The proposed approach outperforms any previously reported method on multiple datasets. The accuracy gain over 2D images is improved by a factor of 1.5. Ablation studies are conducted to demonstrate the significance of each component.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Alperovich, A., Johannsen, O., Strecke, M., Goldluecke, B.: Light field intrinsics with a deep encoder-decoder network. In: CVPR (2018)
Bolles, R.C., Baker, H.H., Marimont, D.H.: Epipolar-plane image analysis: an approach to determining structure from motion. IJCV 1(1), 7–55 (1987). https://doi.org/10.1007/BF00128525
Chen, C., Lin, H., Yu, Z., Bing Kang, S., Yu, J.: Light field stereo matching using bilateral statistics of surface cameras. In: CVPR (2014)
Chen, J., Hou, J., Chau, L.P.: Light field denoising via anisotropic parallax analysis in a CNN framework. IEEE Signal Process. Lett. 25(9), 1403–1407 (2018)
Cho, Y., Bianchi-Berthouze, N., Marquardt, N., Julier, S.J.: Deep thermal imaging: proximate material type recognition in the wild through deep learning of spatial surface temperature patterns. In: Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems (2018)
Cimpoi, M., Maji, S., Vedaldi, A.: Deep filter banks for texture recognition and segmentation. In: CVPR (2015)
DeGol, J., Golparvar-Fard, M., Hoiem, D.: Geometry-informed material recognition. In: CVPR (2016)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR (2016)
Heber, S., Yu, W., Pock, T.: Neural EPI-volume networks for shape from light field. In: ICCV (2017)
Honauer, K., Johannsen, O., Kondermann, D., Goldluecke, B.: A dataset and evaluation methodology for depth estimation on 4D light fields. In: Asian Conference on Computer Vision (2016)
Jeon, H.G., et al.: Accurate depth map estimation from a lenslet light field camera. In: CVPR (2015)
Johannsen, O., Sulc, A., Goldluecke, B.: What sparse light field coding reveals about scene structure. In: CVPR (2016)
Lu, F., He, L., You, S., Chen, X., Hao, Z.: Identifying surface BRDF from a single 4-D light field image via deep neural network. IEEE J. Sel. Top. Signal Process. 11(7), 1047–1057 (2017)
Lu, Z., Yeung, H.W.F., Qu, Q., Chung, Y.Y., Chen, X., Chen, Z.: Improved image classification with 4D light-field and interleaved convolutional neural network. Multimedia Tools Appl. 78(20), 29211–29227 (2018). https://doi.org/10.1007/s11042-018-6597-x
Qi, X., Xiao, R., Li, C.G., Qiao, Y., Guo, J., Tang, X.: Pairwise rotation invariant co-occurrence local binary pattern. IEEE TPAMI 36(11), 2199–2213 (2014)
Reddy, B.S., Chatterji, B.N.: An FFT-based technique for translation, rotation, and scale-invariant image registration. IEEE TIP 5(8), 1266–1271 (1996)
Schwartz, G., Nishino, K.: Material recognition from local appearance in global context. arXiv preprint arXiv:1611.09394 (2016)
Sheng, H., Zhang, S., Cao, X., Fang, Y., Xiong, Z.: Geometric occlusion analysis in depth estimation using integral guided filter for light-field image. IEEE TIP 26(12), 5758–5771 (2017)
Shin, C., Jeon, H.G., Yoon, Y., So Kweon, I., Joo Kim, S.: EPINET: a fully-convolutional neural network using epipolar geometry for depth from light field images. In: CVPR (2018)
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. In: ICLR (2015)
Sonnemann, T., Ulloa Hung, J., Hofman, C.: Mapping indigenous settlement topography in the Caribbean using drones. Remote Sens. 8(10), 791 (2016)
Stone, H.S., Orchard, M.T., Chang, E.C., Martucci, S.A.: A fast direct Fourier-based algorithm for subpixel registration of images. IEEE Trans. Geosci. Remote Sens. 39(10), 2235–2243 (2001)
Storath, M., Weinmann, A.: Fast median filtering for phase or orientation data. IEEE TPAMI 40(3), 639–652 (2018)
Wang, T.C., Efros, A.A., Ramamoorthi, R.: Occlusion-aware depth estimation using light-field cameras. In: ICCV (2015)
Wang, T.C., Zhu, J.Y., Hiroaki, E., Chandraker, M., Efros, A.A., Ramamoorthi, R.: A 4D light-field dataset and CNN architectures for material recognition. In: ECCV (2016)
Wang, T.C., Zhu, J.Y., Kalantari, N.K., Efros, A.A., Ramamoorthi, R.: Light field video capture using a learning-based hybrid imaging system. ACM TOG 36(4), 1–13 (2017)
Wang, Y., Liu, F., Zhang, K., Hou, G., Sun, Z., Tan, T.: LFNet: a novel bidirectional recurrent convolutional neural network for light-field image super-resolution. IEEE TIP 27(9), 4274–4286 (2018)
Wanner, S., Goldluecke, B.: Reconstructing reflective and transparent surfaces from epipolar plane images. In: German Conference on Pattern Recognition (2013)
Weinmann, M., Gall, J., Klein, R.: Material classification based on training data synthesized using a BTF database. In: ECCV (2014)
Wing Fung Yeung, H., Hou, J., Chen, J., Ying Chung, Y., Chen, X.: Fast light field reconstruction with deep coarse-to-fine modeling of spatial-angular clues. In: ECCV (2018)
Wu, G., Zhao, M., Wang, L., Dai, Q., Chai, T., Liu, Y.: Light field reconstruction using deep convolutional network on EPI. In: CVPR (2017)
Xue, J., Zhang, H., Dana, K., Nishino, K.: Differential angular imaging for material recognition. In: CVPR (2017)
Zhang, H., Dana, K., Nishino, K.: Reflectance hashing for material recognition. In: CVPR (2015)
Zhao, C., Sun, L., Stolkin, R.: A fully end-to-end deep learning approach for real-time simultaneous 3D reconstruction and material recognition. In: 2017 18th International Conference on Advanced Robotics (2017)
Zhao, S., Chen, Z.: Light field image coding via linear approximation prior. In: ICIP (2017)
Acknowledgement
Yuxing Han is the corresponding author. This work was supported by the Natural Science Foundation of China (Project Number 61521002) and Shenzhen International Collaborative Research Project (Grant GJHZ20180929151604875).
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
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Guo, B., Wen, J., Han, Y. (2020). Deep Material Recognition in Light-Fields via Disentanglement of Spatial and Angular Information. 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_39
Download citation
DOI: https://doi.org/10.1007/978-3-030-58586-0_39
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)