Journal of the Institute of Convergence Signal Processing (융합신호처리학회논문지)

The Korea Institute of Convergence Signal Processing (한국융합신호처리학회)
권호 표지

A Review of 3D Object Tracking Methods Using Deep Learning

딥러닝 기술을 이용한 3차원 객체 추적 기술 리뷰

  • Park, Hanhoon (Department of Electronic Engineering, Pukyong National University)
  • 박한훈 (부경대학교 전자공학과)
  • Received : 2021.02.17
  • Accepted : 2021.03.29
  • Published : 2021.03.31
Download PDF
⟨ Previous Next ⟩

Abstract

Accurate 3D object tracking with camera images is a key enabling technology for augmented reality applications. Motivated by the impressive success of convolutional neural networks (CNNs) in computer vision tasks such as image classification, object detection, image segmentation, recent studies for 3D object tracking have focused on leveraging deep learning. In this paper, we review deep learning approaches for 3D object tracking. We describe key methods in this field and discuss potential future research directions.

카메라 영상을 이용한 3차원 객체 추적 기술은 증강현실 응용 분야를 위한 핵심 기술이다. 영상 분류, 객체 검출, 영상 분할과 같은 컴퓨터 비전 작업에서 CNN(Convolutional Neural Network)의 인상적인 성공에 자극 받아, 3D 객체 추적을 위한 최근의 연구는 딥러닝(deep learning)을 활용하는 데 초점을 맞추고 있다. 본 논문은 이러한 딥러닝을 활용한 3차원 객체 추적 방법들을 살펴본다. 딥러닝을 활용한 3차원 객체 추적을 위한 주요 방법들을 설명하고, 향후 연구 방향에 대해 논의한다.

Keywords

Acknowledgement

이 논문은 2018년도 정부(교육부)의 재원으로 한국연구재단의 지원을 받아 수행된 기초연구사업임(No. 2018R1D1A1B07045650).

References

  1. A. Dey, M. Billinghurst, R. W. Lindeman, and J. E. Swan, "A systematic review of 10 years of augmented reality usability studies: 2005 to 2014," Front. Robot. AI, vol. 5, article 37, 2018.
  2. K.-M. Lee and J.-I. Kim, "Design and implementation of hybrid VR lock system by Arduino control," The Journal of Korea Institute of Signal Processing and Systems, vol. 15, no. 3, pp. 97-103, 2014.
  3. Y. Wu, F. Tang, and H. Li, "Image-based camera localization: an overview," Visual Computing for Industry, Biometric, and Art, vol. 1, article number: 8, 2018.
  4. V. A. Knyaz, O. Vygolov, V. V. Kniaz, Y. Vizilter, and V. Gorbatsevich, "Deep learning of convolutional auto-encoder for image matching and 3D object reconstruction in the infrared range," Proc. of ICCVW, pp. 2155-2164, 2017.
  5. E. Marchand, H. Uchiyama, and F. Spindler, "Pose estimation for augmented reality: a hands-on survey," IEEE Transactions on Visualization and Computer Graphics, vol. 22, no. 12, pp. 2633-2651, 2016. https://doi.org/10.1109/TVCG.2015.2513408
  6. H. Park and J.-I. Park, "Recent trends and analysis on AR technology - focused on 3D object tracking methods," Proc. of The Korean Institute of Broadcast and Media Engineers Summer Conference, pp. 299-300, 2018.
  7. P. Han and G. Zhao, "A review of edge-based 3D tracking of rigid objects," Virtual Reality & Intelligent Hardware, vol. 1, no. 6, pp. 580-596, 2019. https://doi.org/10.1016/j.vrih.2019.10.001
  8. Y. Shavit and R. Ferens, "Introduction to camera pose estimation with deep learning," arXiv preprint arXiv:1907.05272, 2019.
  9. R. Hartley and A. Zisserman, Multiple View Geometry, 2nd Ed., Cambridge University Press, 2003.
  10. B. Wang, F. Zhong, and X. Qin, "Pose optimization in edge distance field for textureless 3D object tracking," Proc. of the Computer Graphics International Conference, article no. 32, 2017.
  11. X. Liu, J. Zhang, X. He, X. Song, and X. Qin, "6DoF pose estimation with object cutout based on a deep autoencoder," Proc. of ISMAR-Adjunct, 2019.
  12. S. Zhang, C. Song, and R. Radkowski, "Setforge - synthetic RGB-D training data generation to support CNN-based pose estimation for augmented reality," Proc. of ISMAR-Adjunct, pp. 227-232, 2019.
  13. S. Shoman, T. Mashita, A. Plopski, P. Ratsamee, Y. Uranishi, and H. Takemura, "Illumination invariant camera localization using synthetic images," Proc. of ISMAR-Adjunct, pp. 143-144, 2018.
  14. J. Rambach, C. Deng, A. Pagani, and D. Stricker, "Learning 6DoF object poses from synthetic single channel images," Proc. of ISMAR-Adjunct, pp. 164-169, 2018.
  15. J. Tobin, R. Fong, A. Ray, J. Schneider, W. Zaremba, and P. Abbeel, "Domain randomization for transferring deep neural networks from simulation to the real world," Proc. of IROS, pp. 23-30, 2017.
  16. K. M. Yi, E. Trulls, V. Lepetit, and P. Fua, "LIFT: learned invariant feature transform," Proc. of ECCV, pp. 467-483, 2016.
  17. G. Pavlakos, X. Zhou, A. Chan, K. G. Derpanis, and K. Daniilidis, "6-DoF object pose from semantic keypoints," Proc. of ICRA, pp. 2011-2018, 2017.
  18. C. B. Choy, J. Gwak, S. Savarese, and M. Chandraker, "Universal correspondence network," Proc. of NIPS, pp. 2414-2422, 2016.
  19. D. DeTone, T. Malisiewicz, and A. Rabinovich, "SuperPoint: self-supervised interest point detection and description," Proc. of CVPRW, 2018.
  20. D. G. Lowe, "Distinctive image features from scale-invariant keypoints," IJCV, vol. 60, no. 2, pp. 91-110, 2004. https://doi.org/10.1023/B:VISI.0000029664.99615.94
  21. P. Wohlhart and V. Lepetit, "Learning descriptors for object recognition and 3D pose estimation," Proc. of CVPR, pp. 3109-3118, 2015.
  22. W. Kehl, F. Milletari, F. Tombari, S. Ilic, and N. Navab, "Deep learning of local RGB-D patches for 3D object detection and 6D pose estimation," Proc. of ECCV, vol. 3, pp. 205-220, 2016.
  23. K. Park, J. Prankl, and M. Vincze, "Mutual hypothesis verification for 6D pose estimation of natural objects," Proc. of ICCVW, pp. 2192-2199, 2017.
  24. H. Zhang and Q. Cao, "Combined holistic and local patches for recovering 6D object pose," Proc. of ICCVW, pp. 2219-2227, 2017.
  25. A. Crivellaro, M. Rad, Y. Verdie, K. M. Yi, P. Fua, and V. Lepetit, "A novel representation of parts for accurate 3D object detection and tracking in monocular images," Proc. of ICCV, pp. 4391-4399, 2015.
  26. B. Tekin, S. N. Sinha, and P. Fua, "Real-time seamless single shot 6D object pose prediction," Proc. of CVPR, pp. 292-301, 2018.
  27. H. Bay, A. Ess, T. Tuytelaars, and L. Van Gool, "Speeded-up robust features (SURF)," Computer Vision and Image Understanding, vol. 110, no. 3, pp. 346-359, 2008. https://doi.org/10.1016/j.cviu.2007.09.014
  28. N.-D. Duong, A. Kacete, C. Sodalie, P.-Y. Richard, and J. Royan, "xyzNet: towards machine learning camera relocalization by using a scene coordinate prediction network," Proc. of ISMAR-Adjunct, pp. 258-263, 2018.
  29. S. Mahendran, H. Ali, and R. Vidal, "3D pose regression using convolutional neural networks," Proc. of ICCVW, pp. 2174-2182, 2017.
  30. M. Garon and J.-F. Lalonde, "Deep 6-DOF tracking," IEEE Trans. on Vis. and Comp. Grap., vol. 23, no. 11, pp. 2410-2418, 2017. https://doi.org/10.1109/TVCG.2017.2734599
  31. O. Akgul, H. I. Penekli, and Y. Genc, "Applying deep learning in augmented reality tracking," Proc. of SITIS, pp. 47-54, 2016.
  32. H. Su, C. R. Qi, Y. Li, and L. J. Guibas, "Render for CNN: viewpoint estimation in images using CNNs trained with rendered 3D model views," Proc. of ICCV, pp. 2686-2694, 2015.
  33. K. Chatfield, K. Simonyan, A. Vedaldi, and A. Zisserman, "Return of the devil in the details: delving deep into convolutional nets," Proc. of BMVC, 2014.
  34. J. Xiao, A. Owens, and A. Torralba, "SUN3D: a database of big spaces reconstructed using SfM and object labels," Proc. of ICCV, pp. 1625-1632, 2013.
  35. T. Sattler, Q. Zhou, M. Pollefeys, Laura Leal-Taixe, "Understanding the limitations of CNN-based absolute camera pose regression," Proc. of CVPR, pp. 3297-3307, 2019.
  36. K. M. Yi, E. Trulls, Y. Ono, V. Lepetit, M. Salzmann, and P. Fua, "Learning to find good correspondences," Proc. of CVPR, pp. 2666-2674, 2018.
  37. S. Ren, K. He, R. B. Girshick, and J. Sun, "Faster R-CNN: towards real-time object detection with region proposal networks," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp. 1137-1149, 2017. https://doi.org/10.1109/TPAMI.2016.2577031
  38. T.-T. Do, M. Cai, T. Pham, and I. Reid, "Deep-6DPose: recovering 6D object pose from a single RGB image," arXiv preprint arXiv:1802.10367, 2018.
  39. T. X. Qing, W. Fan, and Z. Y. Tao, "Camera pose estimation method based on deep neural network," Proc. of ICDLT, pp. 85-90, 2019.
  40. M Bui, C. Baur, N. Navab, S. Ilic, and S. Albarqouni, "Adversarial networks for camera pose regression and refinement," Proc. of ICCVW, pp. 3778-3787, 2019.
  41. J. R. Rambach, A. Tewari, A. Pagani, and D. Stricker, "Learning to fuse: a deep learning approach to visual-inertial camera pose estimation," Proc. of ISMAR, pp. 71-76, 2016.
  42. V. A. Prisacariu, O. Kahler, D. W. Murray, and I. D. Reid, "Real-time 3D tracking and reconstruction on mobile phones," IEEE Trans. on Vis. and Comp. Grap., vol. 21, no. 5, pp. 557-570, 2015. https://doi.org/10.1109/TVCG.2014.2355207