Skip to main content
SpringerLink
Log in

RelMobNet: End-to-End Relative Camera Pose Estimation Using a Robust Two-Stage Training

  • Conference paper
  • First Online:
Computer Vision – ECCV 2022 Workshops (ECCV 2022)

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 13806))

Included in the following conference series:

Abstract

Relative camera pose estimation, i.e. estimating the translation and rotation vectors using a pair of images taken in different locations, is an important part of systems in augmented reality and robotics. In this paper, we present an end-to-end relative camera pose estimation network using a siamese architecture that is independent of camera parameters. The network is trained using the Cambridge Landmarks data with four individual scene datasets and a dataset combining the four scenes. To improve generalization, we propose a novel two-stage training that alleviates the need of a hyperparameter to balance the translation and rotation loss scale. The proposed method is compared with one-stage training CNN-based methods such as RPNet and RCPNet and demonstrate that the proposed model improves translation vector estimation by 16.11%, 28.88%, and 52.27% on the Kings College, Old Hospital, and St Marys Church scenes, respectively. For proving texture invariance, we investigate the generalization of the proposed method augmenting the datasets to different scene styles, as ablation studies, using generative adversarial networks. Also, we present a qualitative assessment of epipolar lines of our network predictions and ground truth poses.

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 89.00
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 119.99
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.

    https://github.com/3dperceptionlab/therobotrix/issues/1.

  2. 2.

    https://www.nikonians.org/reviews/fov-tables.

  3. 3.

    https://github.com/colmap/colmap/issues/1229.

References

  1. Bailo, O., Rameau, F., Joo, K., Park, J., Bogdan, O., Kweon, I.S.: Efficient adaptive non-maximal suppression algorithms for homogeneous spatial keypoint distribution. Pattern Recogn. Lett. 106, 53–60 (2018)

    Article  Google Scholar 

  2. Bay, H., Ess, A., Tuytelaars, T., Van Gool, L.: Speeded-up robust features (surf). Comput. Vis. Image Underst. 110(3), 346–359 (2008)

    Article  Google Scholar 

  3. Brachmann, E., et al.: Dsac-differentiable ransac for camera localization. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 6684–6692 (2017)

    Google Scholar 

  4. Bradski, G.: The OpenCV library. Dr. Dobb’s J. Softw. Tools (2000)

    Google Scholar 

  5. Chen, K., Snavely, N., Makadia, A.: Wide-baseline relative camera pose estimation with directional learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 3258–3268 (2021)

    Google Scholar 

  6. Dusmanu, M., et al.: D2-net: a trainable CNN for joint detection and description of local features. arXiv preprint arXiv:1905.03561 (2019)

  7. En, S., Lechervy, A., Jurie, F.: Rpnet: an end-to-end network for relative camera pose estimation. In: Proceedings of the European Conference on Computer Vision (ECCV) Workshops, pp. 0–0 (2018)

    Google Scholar 

  8. Fischler, M.A., Bolles, R.C.: Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. Commun. ACM 24(6), 381–395 (1981)

    Article  MathSciNet  Google Scholar 

  9. Graziani, M., Lompech, T., Müller, H., Depeursinge, A., Andrearczyk, V.: On the scale invariance in state of the art CNNs trained on imagenet. Mach. Learn. Knowl. Extraction 3(2), 374–391 (2021)

    Article  Google Scholar 

  10. Hartley, R., Zisserman, A.: Multiple view geometry in computer vision (cambridge university, 2003). C1 C3 2 (2013)

    Google Scholar 

  11. Hartley, R.I.: In defense of the eight-point algorithm. IEEE Trans. Pattern Anal. Mach. Intell. 19(6), 580–593 (1997)

    Article  Google Scholar 

  12. Howard, A., et al.: Searching for mobilenetv3. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 1314–1324 (2019)

    Google Scholar 

  13. Hwang, K., Cho, J., Park, J., Har, D., Ahn, S.: Ferrite position identification system operating with wireless power transfer for intelligent train position detection. IEEE Trans. Intell. Transp. Syst. 20(1), 374–382 (2018)

    Article  Google Scholar 

  14. Kendall, A., Cipolla, R.: Geometric loss functions for camera pose regression with deep learning. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 5974–5983 (2017)

    Google Scholar 

  15. Kendall, A., Grimes, M., Cipolla, R.: Posenet: a convolutional network for real-time 6-dof camera relocalization. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 2938–2946 (2015)

    Google Scholar 

  16. Kim, S., Kim, I., Vecchietti, L.F., Har, D.: Pose estimation utilizing a gated recurrent unit network for visual localization. Appl. Sci. 10(24), 8876 (2020)

    Article  Google Scholar 

  17. Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)

  18. Lee, S., Lee, J., Jung, H., Cho, J., Hong, J., Lee, S., Har, D.: Optimal power management for nanogrids based on technical information of electric appliances. Energy Build. 191, 174–186 (2019)

    Article  Google Scholar 

  19. Lowe, D.G.: Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vision 60(2), 91–110 (2004)

    Article  Google Scholar 

  20. Melekhov, I., Ylioinas, J., Kannala, J., Rahtu, E.: Relative camera pose estimation using convolutional neural networks. In: Blanc-Talon, J., Penne, R., Philips, W., Popescu, D., Scheunders, P. (eds.) ACIVS 2017. LNCS, vol. 10617, pp. 675–687. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-70353-4_57

    Chapter  Google Scholar 

  21. Mina, R.: fast-neural-style: Fast style transfer in pytorch! (2018). https://github.com/iamRusty/fast-neural-style-pytorch

  22. Moraes, C., Myung, S., Lee, S., Har, D.: Distributed sensor nodes charged by mobile charger with directional antenna and by energy trading for balancing. Sensors 17(1), 122 (2017)

    Article  Google Scholar 

  23. Nistér, D.: An efficient solution to the five-point relative pose problem. IEEE Trans. Pattern Anal. Mach. Intell. 26(6), 756–770 (2004)

    Article  Google Scholar 

  24. Paszke, A., et al.: Adaptiveavgpool2d. https://pytorch.org/docs/stable/generated/torch.nn.AdaptiveAvgPool2d.html

  25. Paszke, A., et al.: Pytorch: an imperative style, high-performance deep learning library. In: Wallach, H., Larochelle, H., Beygelzimer, A., d’Alché-Buc, F., Fox, E., Garnett, R. (eds.) Advances in Neural Information Processing Systems 32, pp. 8024–8035. Curran Associates, Inc. (2019). https://papers.neurips.cc/paper/9015-pytorch-an-imperative-style-high-performance-deep-learning-library.pdf

  26. Philbin, J., Chum, O., Isard, M., Sivic, J., Zisserman, A.: Object retrieval with large vocabularies and fast spatial matching. In: 2007 IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–8. IEEE (2007)

    Google Scholar 

  27. Poursaeed, O., et al.: Deep fundamental matrix estimation without correspondences. In: Leal-Taixé, L., Roth, S. (eds.) ECCV 2018. LNCS, vol. 11131, pp. 485–497. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-11015-4_35

    Chapter  Google Scholar 

  28. Raguram, R., Frahm, J.-M., Pollefeys, M.: A comparative analysis of RANSAC techniques leading to adaptive real-time random sample consensus. In: Forsyth, D., Torr, P., Zisserman, A. (eds.) ECCV 2008. LNCS, vol. 5303, pp. 500–513. Springer, Heidelberg (2008). https://doi.org/10.1007/978-3-540-88688-4_37

    Chapter  Google Scholar 

  29. Rublee, E., Rabaud, V., Konolige, K., Bradski, G.: Orb: an efficient alternative to sift or surf. In: 2011 International Conference on Computer Vision, pp. 2564–2571. IEEE (2011)

    Google Scholar 

  30. Sarlin, P.E., DeTone, D., Malisiewicz, T., Rabinovich, A.: Superglue: learning feature matching with graph neural networks. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 4938–4947 (2020)

    Google Scholar 

  31. Schönberger, J.L., Frahm, J.M.: Structure-from-motion revisited. In: Conference on Computer Vision and Pattern Recognition (CVPR) (2016)

    Google Scholar 

  32. Seo, M., Vecchietti, L.F., Lee, S., Har, D.: Rewards prediction-based credit assignment for reinforcement learning with sparse binary rewards. IEEE Access 7, 118776–118791 (2019)

    Article  Google Scholar 

  33. Sun, J., Shen, Z., Wang, Y., Bao, H., Zhou, X.: Loftr: detector-free local feature matching with transformers. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 8922–8931 (2021)

    Google Scholar 

  34. Wynn, K.: pyquaternion (2020). https://github.com/KieranWynn/pyquaternion

  35. Yang, C., Liu, Y., Zell, A.: Rcpnet: deep-learning based relative camera pose estimation for uavs. In: 2020 International Conference on Unmanned Aircraft Systems (ICUAS), pp. 1085–1092. IEEE (2020)

    Google Scholar 

  36. Yew, Z.J., Lee, G.H.: Regtr: end-to-end point cloud correspondences with transformers. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 6677–6686 (2022)

    Google Scholar 

Download references

Acknowledgement

This work was supported by the Institute for Information communications Technology Promotion (IITP) grant funded by the Korean government (MSIT) (No. 2020-0-00440, Development of Artificial Intelligence Technology that continuously improves itself as the situation changes in the real world).

Author information

Authors and Affiliations

  1. Division of Future Vehicle, KAIST, Daejeon, South Korea

    Praveen Kumar Rajendran

  2. The Robotics Program, KAIST, Daejeon, South Korea

    Sumit Mishra

  3. Data Science Group, Institute for Basic Science, Daejeon, South Korea

    Luiz Felipe Vecchietti

  4. The CCS Graduate School of Mobility, KAIST, Daejeon, South Korea

    Dongsoo Har

Authors
  1. Praveen Kumar Rajendran
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Sumit Mishra
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Luiz Felipe Vecchietti
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Dongsoo Har
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Dongsoo Har .

Editor information

Editors and Affiliations

  1. IBM Research - MIT-IBM Watson AI Lab, Massachusetts, USA

    Leonid Karlinsky

  2. Technion – Israel Institute of Technology, Haifa, Israel

    Tomer Michaeli

  3. Kyoto University, Kyoto, Japan

    Ko Nishino

Rights and permissions

Reprints and permissions

Copyright information

© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Rajendran, P.K., Mishra, S., Vecchietti, L.F., Har, D. (2023). RelMobNet: End-to-End Relative Camera Pose Estimation Using a Robust Two-Stage Training. In: Karlinsky, L., Michaeli, T., Nishino, K. (eds) Computer Vision – ECCV 2022 Workshops. ECCV 2022. Lecture Notes in Computer Science, vol 13806. Springer, Cham. https://doi.org/10.1007/978-3-031-25075-0_18

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-25075-0_18

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-25074-3

  • Online ISBN: 978-3-031-25075-0

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Search

Navigation