Kybernetika 62 no. 1, 133-146, 2026

Automatic pose estimation from rigid and partial object imagery

Mehmet Akif AlperDOI: 10.14736/kyb-2026-1-0133

Abstract:

3D pose estimation algorithms have been the subject of widely studied research topic due to problems related to their reliability and precision in related applications. Despite numerous studies by researchers to attempt efficient solutions to application related problems, many proposed methods still not submit sufficiently recover estimates for practical, real-world scenarios in the field of Computer Vision. Therefore, we made extensive study and presented an innovative and practical method that enables a cheap and practical solution by integrating information from both depth and color cameras. Outlier points can impact the pose estimations. We additionally implemented outlier rejection method due to outliers coming from depth to color point projection. After applying and evaluating our proposed algorithm on a public dataset for pose estimation problem, we have shown that it significantly enhances the robustness and accuracy of pose estimation in six degrees of freedom (6-DoF).

Keywords:

transformation, point cloud, projection, rotation, translation

Classification:

68T10, 68T45

paper.pdf

References:

  1. F. Aghili and Ch.-Y. Su: Robust relative navigation by integration of ICP and adaptive Kalman filter using laser scanner and IMU. IEEE/ASME Trans. Mechatronics 2+ (2016), 2015-2026.   DOI:10.1109/tmech.2016.2547905
  2. M. A. Alper, J. Goudreau and M. Daniel: Pose and Optical Flow Fusion (POFF) for accurate tremor detection and quantification. Biocybernet. Biomedical Engrg. 40 (2020), 468-481.   DOI:10.1016/j.bbe.2020.01.009
  3. P. Biber and W. Strasser: The normal distributions transform: a new approach to laser scan matching. In: Proc.2003 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2003) (Cat. No.03CH37453).   DOI:10.1109/iros.2003.1249285
  4. T. Le, L. Hamilton and A. Torralba: Benchmarking convolutional neural networks for object segmentation and pose estimation. On: IEEE Conference Publication, IEEE Applied Imagery Pattern Recognition Workshop 2017.   DOI:10.1109/AIPR.2017.8457942
  5. G. Blais and M. D. Levine: Registering multiview range data to create 3D computer objects. IEEE Trans. Pattern Analysis Machine Intell. 17 (1995), 820-824.   DOI:10.1109/34.400574
  6. E. Brachmann, F. Michel, A. Krull, M. J. Yang, S. Gumhold, C. Rother and Carsten: Uncertainty-driven 6D pose estimation of objects and scenes from a single RGB image. IEEE Xplore (2016), 3364-3372.   DOI:10.1109/CVPR.2016.366
  7. M. Brunot, A. Janot, P. C. Young and F. Carrillo: An instrumental variable method for robot identification based on time variable parameter estimation. Kybernetika 54 (2018), 1.   DOI:10.14736/kyb-2018-1-0202
  8. S. Chen, L. Liang, W. Luming and H. Foroosh: 3D pose tracking with multi-template warping and SIFT correspondences. IEEE Trans. Circuits Systems Video Technol. (2015).   DOI:10.1109/tcsvt.2015.2452782
  9. Y. Chen and G. Medioni: Object modelling by registration of multiple range images. Image Vision Comput. 10 (1992), 145-155.   DOI:10.1016/0262-8856(92)90066-c
  10. M. Delavari, A. H. Foruzan and Y.-W. Chen: Accurate point correspondences using a modified coherent point drift algorithm. Biomedical Signal Process. Control 52 (2019), 429-444.   DOI:10.1016/j.bspc.2017.02.009
  11. J. Deng, Y. Pan, T. Yao, W. Zhou, H. Li and T. Mei: Single shot video object detector. IEEE Trans. Multimedia 23 (2021), 846-858.   DOI:10.1109/TMM.2020.2990070
  12. D. Ding, R. A. Cooper, P. F. Pasquina and L. Fici-Pasquina: Sensor technology for smart homes. Maturitas 69 (2011), 131-136.   DOI:10.1016/j.maturitas.2011.03.016
  13. W. Gao and R. Tedrake: FilterReg: Robust and Efficient Probabilistic Point-Set Registration Using Gaussian Filter and Twist Parameterization. Massachusetts Institute of Technology, 2019.   DOI:10.1109/cvpr.2019.01135
  14. L. A. Giefer, J. D. Castellanos, M. M. Babr and M. Freitag: Deep learning-based pose estimation of apples for inspection in logistic centers using single-perspective imaging. Processes 7 (2019), 424.   DOI:10.3390/pr7070424
  15. Z. He, Z. Jiang, X. Zhao, S. Zhang and Ch. Wu: Sparse template-based 6-D pose estimation of metal parts using a monocular camera. IEEE Trans. Industrial Electronics 67 (2020), 390-401.   DOI:10.1109/tie.2019.2897539
  16. H. K. Hong, H. Yu and B.-H. Lee: Regeneration of normal distributions transform for target lattice based on fusion of truncated Gaussian components. IEEE Robotics Automation Lett. 4 (2019), 684-691.   DOI:10.1109/lra.2019.2891493
  17. G. Hua, L. Li and S. Liu: Multipath affinage stacked—hourglass networks for human pose estimation. Frontiers Computer Sci. 14 (2020).   DOI:10.1007/s11704-019-8266-2
  18. A. Kendall, M. Grimes and R. Cipolla: PoseNet: A convolutional network for real-time 6-DOF camera relocalization. In: 2015 IEEE International Conference on Computer Vision (ICCV).   DOI:10.1109/iccv.2015.336
  19. L. Kevin, B. Liefeng and D. Fox: Unsupervised feature learning for 3D scene labeling. In: IEEE International Conference on Robotics and Automation (ICRA), 2014.   CrossRef
  20. N. A. Kilyen, R. F. Lemnariu, I. Muntean and G. D. Mois: An Indoor Localization System for Automotive Driving Competitions, Int. J. Computers Commun. Control 19 (2024), 1.   DOI:10.15837/ijccc.2024.1.6030
  21. D. W. Leng and W. D. Sun: Contour-based iterative pose estimation of 3D rigid object. IET Computer Vision 5 (2011), 291.   DOI:10.1049/iet-cvi.2010.0098
  22. M. Li, T. Oncel, V. Ashok and Ch. Rama: Fast directional chamfer matching. Pennsylvania State University, 2010.   DOI:10.1109/cvpr.2010.5539837
  23. D. Liu, S. Arai, J. Miao, J.Kinugawa, Z. Wang and K. Kosuge: Point pair feature-based pose estimation with multiple edge appearance models (PPF-MEAM) for robotic bin picking. Sensors 18 (2018), 2719.   DOI:10.3390/s18082719
  24. P. Liu, M. R. Lyu, I. King and J. Xu: SelFlow: Self-supervised learning of optical flow. CVPR (2019).   DOI:10.1109/cvpr.2019.00470
  25. K. Liu, W. Wang, K. Thiagalingam and J. Wang: Dynamic vehicle detection with sparse point clouds based on PE-CPD. Institute of Electrical and Electronics Engineers 20 (2019), 1964-1977.   DOI:10.1109/tits.2018.2857510
  26. T. Liu, J. Zheng, Z. Wang, Z. Huang and Y. Chen: Composite clustering normal distribution transform algorithm. Int. J. Advanced Robotic Systems, SAGE Publish. 17 (2020).   DOI:10.1177/1729881420912142
  27. A. Myronenko and S. Xubo: Point Set Registration: Coherent Point Drift. IEEE Trans. Pattern Anal. Machine Intell. 32 (2010), 2262-2275.   DOI:10.1109/tpami.2010.46
  28. R. Opromolla, G. Fasano, G. Rufino and M. Grassi: A model-based 3D template matching technique for pose acquisition of an uncooperative space object. Sensors 15 (2015), 6260-6382.   DOI:10.3390/s150306360
  29. K. Picos, V. H. Diaz-Ramirez, V. Kober, A. S. Montemayor and J. J. Pantrigo: Accurate three-dimensional pose recognition from monocular images using template matched filtering. Optical Engrg. 55 (2016).   DOI:10.1117/1.oe.55.6.063102
  30. T. G. Phillips and P. R. McAree: An evidence-based approach to object pose estimation from LiDAR measurements in challenging environments. J. Field Robotics 35 (2018), 921-936.   DOI:10.1002/rob.21788
  31. D. Presnov, M. Lambers and A. Kolb: Robust range camera pose estimation for mobile online scene reconstruction. IEEE Sensors J. 18 (2018), 2903-2915.   DOI:10.1109/jsen.2018.2801878
  32. S. Quan, J. Ma, Jie, F. Hu, B. Fang and T. Ma: Local voxelized structure for 3D binary feature representation and robust registration of point clouds from low-cost sensors. Inform. Sci. 444 (2018), 153-171.   DOI:10.1016/j.ins.2018.02.070
  33. J. Schlobohm, A. Pösch, E. Reithmeier and B. Rosenhahn: Improving contour based pose estimation for fast 3D measurement of free form objects. Measurement 92 (2016), 79-82.   DOI:10.1016/j.measurement.2016.05.093
  34. A. Singh, J. Sha, K. S. Narayan, S. Karthik, T. Achim and A. Pieter: BigBIRD: A large-scale 3D database of object instances. In: International Conference on Robotics and Automation, 2014.   DOI:10.1109/icra.2014.6906903
  35. K.-T. Song, Ch.-H. Wu and S.-Y. Jiang: CAD-based Pose estimation design for random bin picking using a RGB-D camera. J. Intell. Robotic Systems 87 (2017), 455-470.   DOI:10.1007/s10846-017-0501-1
  36. X. Teng, Q. Yu, Qifeng, J. Luo, G. Wang and X. Zhang: Aircraft pose estimation based on geometry structure features and line correspondences. Sensors 19 (2019), 2165.   DOI:10.3390/s19092165
  37. M. I. Thorbjorn G. B. Anders, N. Krüger and D. Kraft: Shape dependency of ICP pose uncertainties in the context of pose estimation systems. Lecture Notes Computer Sci. (2015), 303-315.   DOI:10.1007/978-3-319-20904-3-28
  38. Ch.-Y. Tsai, K.-J. Hsu and H. Nisar: Efficient model-based object pose estimation based on multi-template tracking and PnP algorithms. Algorithms 11 (2018), 122.   DOI:10.3390/a11080122
  39. B. Wang, F. Zhong and X. Qin: Robust edge-based 3D object tracking with direction-based pose validation. Multimedia Tools Appl. 78 (2018), 12307-12331.   DOI:10.1007/s11042-018-6727-5
  40. T. Yang, Q. Zhao, X. Wang and Q. Zhou: Sub-pixel chessboard corner localization for camera calibration and pose estimation. Appl. Sci. 8 (2018), 2118.   DOI:10.3390/app8112118
  41. A. Zeng, K.-T. Yu, S. Song, D. Suo, E. Walker, A. Rodriguez and J. Xiao: Multi-view self-supervised deep learning for 6D pose estimation in the Amazon Picking Challenge. IEEE Xplore (2017), 1386-1383.   DOI:10.1109/ICRA.2017.7989165
  42. R. Zhao, H. Ali and P. van der Smagt: Two-stream RNN/CNN for action recognition in 3D videos. In: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017.   DOI:10.1109/iros.2017.8206288
  43. X. Zhang, Z. Jiang, H. Zhang and Q. Wei: Vision-based pose estimation for textureless space objects by contour points matching. IEEE Trans. Aerospace Electron. Systems 54 (2018), 2342-2355.   DOI:10.1109/taes.2018.2815879