Skip to main content
SpringerLink
Log in

Fast Light Field Disparity Estimation via a Parallel Filtered Cost Volume Approach

  • Conference paper
  • First Online:
Book cover Computer Vision – ACCV 2018 (ACCV 2018)

Part of the book series: Lecture Notes in Computer Science ((LNIP,volume 11362))

Included in the following conference series:

  • 2079 Accesses

Abstract

The Guided Light Field Cost Volume (GLFCV) is a light field disparity estimation algorithm designed for (GPU) parallelization by refactoring the process, such that costly optimizations that combine and refine depth maps are simplified. The algorithm involves shearing the light field over a range of disparities and computing a cost volume for each sheared sub-aperture image. A guided filter is then run on the computed cost for each disparity. For the final disparity estimate, each pixel is assigned disparity associated with the lowest filtered cost. Execution time and accuracy were evalulated on Lytro Illum imagery and also on synthetic light fields from the HCI 4D light field benchmark. The approach presented executes in four seconds (on an NVIDIA Titan XP), with only 10% of pixels in the disparity estimate deviating from ground truth. This compares favourably to existing accurate approaches, whose execution time is in the order of minutes.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution
Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 39.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 54.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

  1. 1.

    Four dimensional light field data can be parameterised as L(x, y, s, t) [9], which is a pair of planar coordinates where for every fixed pair of (s, t), varying the (x, y) coordinates gives a different pinhole view (or sub-aperture image) of the scene. This parameterisation is used throughout.

  2. 2.

    Available at: http://hci-lightfield.iwr.uni-heidelberg.de.

References

  1. Blender Online Community: Blender - a 3D modelling and rendering package. Blender Foundation, Blender Institute, Amsterdam (2017). http://www.blender.org

  2. Bradski, G.: The OpenCV library. Dr. Dobb’s Journal of Software Tools (2000)

    Google Scholar 

  3. Dansereau, D., Bruton, L.: Gradient-based depth estimation from 4D light fields. In: Proceedings of the 2004 International Symposium on Circuits and Systems, ISCAS 2004, vol. 3, pp. III-549. IEEE (2004)

    Google Scholar 

  4. He, K., Sun, J.: Fast guided filter. CoRR abs/1505.00996 (2015). http://arxiv.org/abs/1505.00996

  5. Honauer, K., Johannsen, O., Kondermann, D., Goldluecke, B.: A dataset and evaluation methodology for depth estimation on 4D light fields. In: Lai, S.-H., Lepetit, V., Nishino, K., Sato, Y. (eds.) ACCV 2016. LNCS, vol. 10113, pp. 19–34. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-54187-7_2

    Chapter  Google Scholar 

  6. Hosni, A., Bleyer, M., Gelautz, M.: Secrets of adaptive support weight techniques for local stereo matching. Comput. Vis. Image Underst. 117(6), 620–632 (2013). https://doi.org/10.1016/j.cviu.2013.01.007

    Article  Google Scholar 

  7. Jeon, H.G., et al.: Accurate depth map estimation from a lenslet light field camera. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1547–1555. IEEE (2015)

    Google Scholar 

  8. Jeon, H.G., et al.: Depth from a light field image with learning-based matching costs. IEEE Trans. Pattern Anal. Mach. Intell. PP(99), 1–14 (2017)

    Google Scholar 

  9. Johannsen, O., et al.: A taxonomy and evaluation of dense light field depth estimation algorithms. In: 2nd Workshop on Light Fields for Computer Vision at CVPR (2017)

    Google Scholar 

  10. Johannsen, O., Sulc, A., Goldluecke, B.: What sparse light field coding reveals about scene structure. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3262–3270 (2016)

    Google Scholar 

  11. Luo, Y., Zhou, W., Fang, J., Liang, L., Zhang, H., Dai, G.: EPI-patch based convolutional neural network for depth estimation on 4D light field. In: Liu, D., Xie, S., Li, Y., Zhao, D., El-Alfy, E.S. (eds.) ICONIP 2017. LNCS, vol. 10636, pp. 642–652. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-70090-8_65

    Chapter  Google Scholar 

  12. Ma, Z., He, K., Wei, Y., Sun, J., Wu, E.: Constant time weighted median filtering for stereo matching and beyond. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 49–56 (2013)

    Google Scholar 

  13. Neri, A., Carli, M., Battisti, F.: A multi-resolution approach to depth field estimation in dense image arrays. In: 2015 IEEE International Conference on Image Processing (ICIP), pp. 3358–3362, September 2015. https://doi.org/10.1109/ICIP.2015.7351426

  14. Ng, R., Levoy, M., Brédif, M., Duval, G., Horowitz, M., Hanrahan, P.: Light field photography with a hand-held plenoptic camera. Comput. Sci. Techn. Rep. CSTR 2(11), 1–11 (2005)

    Google Scholar 

  15. NVIDIA Corporation: CUDA parallel computing platform (2016). http://www.nvidia.com/object/cuda_home_new.html

  16. Raj, A.S., Lowney, M., Shah, R.: Light-field database creation and depth estimation (2016). https://stanford.edu/class/ee367/Winter2016/Lowney_Shah_Sunder_Raj_Report.pdf, computational Imaging and Display Report (Winter 2016)

  17. Shin, C., Jeon, H.G., Yoon, Y., Kweon, I.S., Kim, S.J.: Epinet: a fully-convolutional neural network using epipolar geometry for depth from light field images. CoRR abs/1804.02379 (2018)

    Google Scholar 

  18. Si, L., Wang, Q.: Dense depth-map estimation and geometry inference from light fields via global optimization. In: Lai, S.-H., Lepetit, V., Nishino, K., Sato, Y. (eds.) ACCV 2016. LNCS, vol. 10113, pp. 83–98. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-54187-7_6

    Chapter  Google Scholar 

  19. Stacey, A.: GPU Acceleration of Light Field Depth Estimation. Bachelor thesis, The University of Queensland, June 2017

    Google Scholar 

  20. Strecke, M., Alperovich, A., Goldluecke, B.: Accurate depth and normal maps from occlusion-aware focal stack symmetry. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2017)

    Google Scholar 

  21. Tao, M.W., Hadap, S., Malik, J., Ramamoorthi, R.: Depth from combining defocus and correspondence using light-field cameras. In: The IEEE International Conference on Computer Vision (ICCV), October 2013

    Google Scholar 

  22. Tao, M.W., Srinivasan, P.P., Malik, J., Rusinkiewicz, S., Ramamoorthi, R.: Depth from shading, defocus, and correspondence using light-field angular coherence. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1940–1948 (2015)

    Google Scholar 

  23. Wang, T.C., Efros, A.A., Ramamoorthi, R.: Occlusion-aware depth estimation using light-field cameras. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 3487–3495 (2015)

    Google Scholar 

  24. Wanner, S., Goldluecke, B.: Globally consistent depth labeling of 4D light fields. In: 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 41–48. IEEE (2012)

    Google Scholar 

  25. Zhang, S., Sheng, H., Li, C., Zhang, J., Xiong, Z.: Robust depth estimation for light field via spinning parallelogram operator. Comput. Vis. Image Underst. 145, 148–159 (2016)

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. The University of Queensland, Brisbane, Australia

    Adam Stacey & Surya Singh

  2. University of Oxford, Oxford, UK

    Will Maddern

Authors
  1. Adam Stacey
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Will Maddern
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Surya Singh
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Adam Stacey .

Editor information

Editors and Affiliations

  1. IIIT Hyderabad, Hyderabad, India

    C. V. Jawahar

  2. ANU, Canberra, ACT, Australia

    Hongdong Li

  3. Simon Fraser University, Burnaby, BC, Canada

    Greg Mori

  4. ETH Zurich, Zurich, Zürich, Switzerland

    Konrad Schindler

Rights and permissions

Reprints and permissions

Copyright information

© 2019 Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Stacey, A., Maddern, W., Singh, S. (2019). Fast Light Field Disparity Estimation via a Parallel Filtered Cost Volume Approach. In: Jawahar, C., Li, H., Mori, G., Schindler, K. (eds) Computer Vision – ACCV 2018. ACCV 2018. Lecture Notes in Computer Science(), vol 11362. Springer, Cham. https://doi.org/10.1007/978-3-030-20890-5_17

Download citation

  • DOI: https://doi.org/10.1007/978-3-030-20890-5_17

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-030-20889-9

  • Online ISBN: 978-3-030-20890-5

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Search

Navigation