предыдущий|следующий

Domain adaptation learning for 3D point clouds： A survey

  • role： First author第一作者
  • Affiliation：

    College of Computer and Data Science, Fuzhou University, Fuzhou 350108, China

  • Email：fanwenhui99@qq.com
  • Introduction：范文辉， 研究方向为三维点云处理。E-mail： fanwenhui99@qq.com
FAN Wenhui1,  
  • Affiliation：

    College of Computer and Data Science, Fuzhou University, Fuzhou 350108, China

LIN Xi1,  
  • Affiliation：

    College of Computer and Data Science, Fuzhou University, Fuzhou 350108, China

LUO Huan1,  
  • Affiliation：

    College of Computer and Data Science, Fuzhou University, Fuzhou 350108, China

GUO Wenzhong1,  
  • role： Corresponding author通信作者
  • Affiliation：

    School of Surveying and Mapping, Information Engineering University, Zhengzhou 450001, China

  • Email：why.scholar@gmail.com
  • Introduction：汪汉云，研究方向为三维点云智能处理与理解。E-mail： why.scholar@gmail.com
WANG Hanyun2*,  
  • Affiliation：

    School of Surveying and Mapping, Information Engineering University, Zhengzhou 450001, China

DAI Chenguang2

реферат

Three-dimensional (3D) point cloud data have been widely used in many fields, such as autonomous driving, robotics, and high-precision mapping. At present, the state-of-the-art deep learning-based methods for 3D point cloud processing are mainly supervised learning methods. The performance of these methods depends heavily on large-scale, high-quality annotated datasets. However, annotating a large-scale, high-quality, category-diverse, and scenario-rich dataset is time-consuming and labor-intensive. In particular, obtaining sufficiently large numbers of samples for model optimization is also quite difficult in some special cases. In addition, 3D point cloud processing models trained on a single device in a special environment are difficult to generalize to different devices and environments. Their generalizability to various devices and environments is limited. Thus, how to reduce dependencies on high-quality annotated 3D point cloud datasets and how to improve the generalizability of current point cloud processing models are important research topics. In recent years, various kinds of impressive and elaborate technologies, such as meta-learning, few-shot learning, transfer learning, self-supervised learning, semisupervised learning, and weakly supervised learning, have been proposed to solve this problem. As an important research branch of transfer learning, domain adaptive learning aims to eliminate differences in feature distributions across domains and promote the generalization ability of deep learning models, thereby providing a novel solution to address this problem effectively. The academic community has conducted preliminary research on domain adaptive learning for point cloud processing. However, the domain adaptive learning field for point clouds still requires in-depth and effective exploration. Consequently, this study systematically summarizes and classifies recent 3D point cloud domain adaptive learning methods into five categories: adversarial learning, cross-modal learning, pseudo-label learning, data alignment, and other kinds of methods. First, we present the mathematical definition of the domain adaptive learning task and depict the chronological overview of the development of different domain adaptive learning methods to provide readers with a clear understanding. Second, we present the general solution for each category of domain adaptive learning methods and summarize the advantages and disadvantages of the current methods for each category. Third, we compare the performance of current methods on three-point cloud processing tasks, including 3D shape classification, 3D object detection, and 3D semantic segmentation. For each task, we also summarize the commonly used datasets and evaluation metrics for an intuitional comparison. Finally, we conclude the advantages and disadvantages of these five categories of methods and discuss future research directions about the 3D point cloud domain adaptive learning.

ключеви́че слова́

remote sensing;3D point cloud;domain adaption learning;adversarial learning;cross-modal learning;pseudo-label learning;data alignment

References

  1. 1.
    Achituve I, Maron H and Chechik G. 2021. Self-supervised learning for domain adaptation on point clouds//2021 IEEE Winter Conference on Applications of Computer Vision (WACV). Waikoloa: IEEE: 123-133
  2. 2.
    Alliegro A, Boscaini D and Tommasi T. 2021. Joint supervised and self-supervised learning for 3D real world challenges//2020 25th International Conference on Pattern Recognition (ICPR). Milan: IEEE: 6718-6725
  3. 3.
    Behley J, Garbade M, Milioto A, Quenzel J, Behnke S, Stachniss C and Gall J. 2019. SemanticKITTI: a dataset for semantic scene understanding of LiDAR sequences//2019 IEEE/CVF International Conference on Computer Vision (ICCV). Seoul: IEEE: 9296-9306
  4. 4.
    Bian Y K, Hui L, Qian J J and Xie J. 2022. Unsupervised domain adaptation for point cloud semantic segmentation via graph matching//2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Kyoto: IEEE: 9899-9904
  5. 5.
    Caesar H, Bankiti V, Lang A H, Vora S, Liong V E, Xu Q, Krishnan A, Pan Y, Baldan G and Beijbom O. 2020. NuScenes: a multimodal dataset for autonomous driving//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE: 11618-11628
  6. 6.
    Cardace A, Spezialetti R, Ramirez P Z, Salti S and Stefano L D. 2021. RefRec: pseudo-labels refinement via shape reconstruction for unsupervised 3D domain adaptation//2021 International Conference on 3D Vision (3DV). London: IEEE: 331-341
  7. 7.
    Chang A X, Funkhouser T, Guibas L, Hanrahan P, Huang Q X, Li Z M, Savarese S, Savva M, Song S R, Su H, Xiao J X, Yi L and Yu F. 2015. ShapeNet: an information-rich 3D model repository. arXiv: 1512.03012
  8. 8.
    Charles R Q, Su H, Kaichun M and Guibas L J. 2017. PointNet: deep learning on point sets for 3D classification and segmentation//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu: IEEE: 77-85
  9. 9.
    Chen C, Fu Z H, Chen Z H, Jin S., Cheng Z W, Jin X Y and Hua X S. 2020. Homm: Higher-order moment matching for unsupervised domain adaptation// Proceedings of the AAAI conference on artificial intelligence. 34(04):3422-3429
  10. 10.
    Chen J J, Kakillioglu B, Ren H and Velipasalar S. 2022. Why discard if you can recycle?: A recycling max pooling module for 3d point cloud analysis//2022 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). New Orleans. IEEE: 559-567.
  11. 11.
    Chiang H Y, Lin Y L, Liu Y C and Hsu W H. 2019. A unified point-based framework for 3D segmentation//2019 International Conference on 3D Vision (3DV). Quebec City: IEEE: 155-163
  12. 12.
    Cortés I, Beltrán J, De La Escalera A and García F. 2022. DALi: domain adaptation in LiDAR point clouds for 3D obstacle detection//2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC). Macau, China: IEEE: 3837-3842
  13. 13.
    Dai A, Chang A X, Savva M, Halber M, Funkhouser T and Niessner M. 2017. ScanNet: richly-annotated 3D reconstructions of indoor scenes//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu: IEEE: 2432-2443
  14. 14.
    Dong Z, Liang F X, Yang B S, Xu Y S, Zang Y F, Li J P, Wang Y, Dai W X, Fan H C, Hyyppä J and Stilla U. 2020. Registration of large-scale terrestrial laser scanner point clouds: a review and benchmark. ISPRS Journal of Photogrammetry and Remote Sensing, 163: 327-342
  15. 15.
    Fan H H, Chang X J, Zhang W Y, Cheng Y, Sun Y and Kankanhalli M. 2022. Self-supervised global-local structure modeling for point cloud domain adaptation with reliable voted pseudo labels//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). New Orleans: IEEE: 6367-6376
  16. 16.
    Fruhwirth-Reisinger C, Opitz M, Possegger H and Bischof H. 2021. FAST3D: flow-aware self-training for 3D object detectors. arXiv: 2110.09355
  17. 17.
    Geiger A, Lenz P and Urtasun R. 2012. Are we ready for autonomous driving? The KITTI vision benchmark suite//2012 IEEE Conference on Computer Vision and Pattern Recognition. Providence: IEEE: 3354-3361
  18. 18.
    Geyer J, Kassahun Y, Mahmudi M, Ricou X, Durgesh R, Chung A S, Hauswald L, Pham V H, Mühlegg M, Dorn S, Fernandez T, Jänicke M, Mirashi S, Savani C, Sturm M, Vorobiov O, Oelker M, Garreis S and Schuberth P. 2020. A2D2: audi autonomous driving dataset. arXiv: 2004.06320
  19. 19.
    Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A and Bengio Y. 2020. Generative adversarial networks. Communications of the ACM, 63(11): 139-144
  20. 20.
    Hackel T, Savinov N, Ladicky L, Wegner J D, Schindler K and Pollefeys M. 2017. Semantic3D.net: a new large-scale point cloud classification benchmark. arXiv: 1704.03847
  21. 21.
    Hu Q Y, Yang B, Xie L H, Rosa S, Guo Y L, Wang Z H, Trigoni N and Markham A. 2020. RandLA-Net: efficient semantic segmentation of large-scale point clouds//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE: 11105-11114
  22. 22.
    Huang J X, Yuan J S and Qiao C M. 2022. Generation for unsupervised domain adaptation: a gan-based approach for object classification with 3D point cloud data//ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Singapore: IEEE: 3753-3757
  23. 23.
    Jaritz M, Gu J Y and Su H. 2019. Multi-view PointNet for 3D scene understanding//2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW). Seoul: IEEE: 3995-4003
  24. 24.
    Jaritz M, Vu T H, de Charette R, Wirbel E and Pérez P. 2020. XMUDA: cross-modal unsupervised domain adaptation for 3D semantic segmentation//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE: 12602-12611
  25. 25.
    Jiang P and Saripalli S. 2021. LiDARNet: a boundary-aware domain adaptation model for point cloud semantic segmentation//2021 IEEE International Conference on Robotics and Automation (ICRA). Xi’an: IEEE: 2457-2464
  26. 26.
    Kang D M, Nam Y, Kyung D and Choi J. 2022. Unsupervised domain adaptation for 3D point clouds by searched transformations. IEEE Access, 10: 56901-56913
  27. 27.
    Li Q, Peng X J, Yan C, Gao P and Hao Q. 2023. Self-ensemling for 3D point cloud domain adaption. arXiv: 2112.05301
  28. 28.
    Liang H X, Fan H H, Fan Z W, Wang Y, Chen T L, Cheng Y and Wang Z Y. 2022. Point cloud domain adaptation via masked local 3D structure prediction//17th European Conference on Computer Vision. Tel Aviv: Springer: 156-172
  29. 29.
    Liu W, Luo Z M, Cai Y Z, Yu Y, Ke Y, Junior J M, Gonçalves W N and Li J. 2021. Adversarial unsupervised domain adaptation for 3D semantic segmentation with multi-modal learning. ISPRS Journal of Photogrammetry and Remote Sensing, 176: 211-221
  30. 30.
    Liu X Y, Qi C R and Guibas L J. 2019. FlowNet3D: learning scene flow in 3D point clouds//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Long Beach: IEEE: 529-537
  31. 31.
    Lu W X, Zhou Y, Wan G W, Hou S H and Song S Y. 2019. L3-Net: towards learning based LiDAR localization for autonomous driving//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Long Beach: IEEE: 6382-6391
  32. 32.
    Luo H, Li L K, Fang L N, Wang H Y, Wang C, Guo W Z and Li J. 2022. Domain adaptation for object classification in point clouds via asymmetrical siamese and conditional adversarial network. IEEE Geoscience and Remote Sensing Letters, 19: 7004605
  33. 33.
    Luo H F, Khoshelham K, Fang L N and Chen C C. 2020. Unsupervised scene adaptation for semantic segmentation of urban mobile laser scanning point clouds. ISPRS Journal of Photogrammetry and Remote Sensing, 169: 253-267
  34. 34.
    Luo X Y, Liu S L, Fu K X, Wang M N and Song Z J. 2021a. A learnable self-supervised task for unsupervised domain adaptation on point clouds. arXiv: 2104.05164
  35. 35.
    Luo Z P, Cai Z, Zhou C Q, Zhang G J, Zhao H Y, Yi S, Lu S J, Li H S, Zhang S H and Liu Z W. 2021b. Unsupervised domain adaptive 3D detection with multi-level consistency//2021 IEEE/CVF International Conference on Computer Vision (ICCV). Montreal: IEEE: 8846-8855
  36. 36.
    Pan S J and Yang Q. 2010. A survey on transfer learning. IEEE Transactions on Knowledge and Data Engineering, 22(10): 1345-1359
  37. 37.
    Peng D, Lei Y J, Li W, Zhang P P and Guo Y L. 2021. Sparse-to-dense feature matching: intra and inter domain cross-modal learning in domain adaptation for 3D semantic segmentation//2021 IEEE/CVF International Conference on Computer Vision (ICCV). Montreal: IEEE: 7088-7097
  38. 38.
    Pinheiro P O. 2018. Unsupervised domain adaptation with similarity learning//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE: 8004-8013
  39. 39.
    Qi C R, Yi L, Su H and Guibas L J. 2017. PointNet++: deep hierarchical feature learning on point sets in a metric space. arXiv: 1706.02413
  40. 40.
    Qin C, You H X, Wang L C, Kuo C C J and Fu Y. 2019. PointDAN: a multi-scale 3D domain adaption network for point cloud representation. arXiv: 1911.02744
  41. 41.
    Rubner Y, Tomasi C and Guibas L J. 2000. The earth mover’s distance as a metric for image retrieval. International Journal of Computer Vision, 40(2): 99-121
  42. 42.
    Saito K, Watanabe K, Ushiku Y and Harada T. 2018. Maximum classifier discrepancy for unsupervised domain adaptation//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE: 3723-3732
  43. 43.
    Saleh K, Abobakr A, Attia M, Iskander J, Nahavandi D, Hossny M and Nahvandi S. 2019. Domain adaptation for vehicle detection from bird’s eye view LiDAR point cloud data//2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW). Seoul: IEEE: 3235-3242
  44. 44.
    Saltori C, Lathuilière S, Sebe N, Ricci E and Galasso F. 2020. SF-UDA3D: source-free unsupervised domain adaptation for LiDAR-based 3D object detection//2020 International Conference on 3D Vision (3DV). Fukuoka: IEEE: 771-780
  45. 45.
    Shen Y F, Yang Y C, Yan M, Wang H, Zheng Y Y and Guibas L. 2022. Domain adaptation on point clouds via geometry-aware implicits//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). New Orleans: IEEE: 7213-7222
  46. 46.
    Shi L W, Yuan Z M, Cheng M, Chen Y P and Wang C. 2022. DFAN: dual-branch feature alignment network for domain adaptation on point clouds. IEEE Transactions on Geoscience and Remote Sensing, 60: 5703412
  47. 47.
    Shi S S, Wang X G and Li H S. 2019. PointRCNN: 3D Object Proposal Generation and Detection From Point Cloud//2019 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Long Beach: IEEE: 770-779
  48. 48.
    Sun P, Kretzschmar H, Dotiwalla X, Chouard A, Patnaik V, Tsui P, Guo J, Zhou Y, Chai Y N, Caine B, Vasudevan V, Han W, Ngiam J, Zhao H, Timofeev A, Ettinger S, Krivokon M, Gao A, Joshi A, Zhang Y, Shlens J, Chen Z F and Anguelov D. 2020. Scalability in perception for autonomous driving: waymo open dataset//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE: 2443-2451
  49. 49.
    Sun Y, Tzeng E, Darrell T and Efros A A. 2019. Unsupervised domain adaptation through self-supervision. arXiv: 1909.11825
  50. 50.
    Takikawa T, Acuna D, Jampani V and Fidler S. 2019. Gated-SCNN: gated shape CNNs for semantic segmentation//2019 IEEE/CVF International Conference on Computer Vision (ICCV). Seoul: IEEE: 5228-5237
  51. 51.
    Tang H, Chen K and Jia K. 2020. Unsupervised domain adaptation via structurally regularized deep clustering//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE: 8722-8732
  52. 52.
    Tang H R, Xu C W and Yang J F. 2021. Bi-adversarial discrepancy minimization for unsupervised domain adaptation on 3D point cloud//2021 International Joint Conference on Neural Networks (IJCNN). Shenzhen: IEEE: 1-8
  53. 53.
    Tarvainen A and Valpola H. 2018. Mean teachers are better role models: weight-averaged consistency targets improve semi-supervised deep learning results. arXiv: 1703.01780
  54. 54.
    Thomas H, Qi C R, Deschaud J E, Marcotegui B, Goulette F and Guibas L. 2019. KPConv: flexible and deformable convolution for point clouds//2019 IEEE/CVF International Conference on Computer Vision (ICCV). Seoul: IEEE: 6410-6419
  55. 55.
    Ulyanov D, Vedaldi A and Lempitsky V. 2017. Instance normalization: The missing ingredient for fast stylization. arXiv:1607.08022.
  56. 56.
    Wang Y, Chen X Y, You Y R, Li L E, Hariharan B, Campbell M, Weinberger K Q and Chao W L. 2020. Train in Germany, test in the USA: making 3D object detectors generalize//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE: 11710-11720
  57. 57.
    Wang Y, Yin J B, Li W, Frossard P, Yang R G and Shen J B. 2022. SSDA3D: semi-supervised domain adaptation for 3D object detection from point cloud. arXiv: 2212.02845
  58. 58.
    Wu B C, Wan A, Yue X Y and Keutzer K. 2018. Squeezeseg: Convolutional neural nets with recurrent crf for real-time road-object segmentation from 3d lidar point cloud.//2018 IEEE international conference on robotics and automation (ICRA). Brisbane: IEEE: 1887-1893 []
  59. 59.
    Wu B C, Zhou X Y, Zhao S C, Yue X Y and Keutzer K. 2019. SqueezeSegV2: improved model structure and unsupervised domain adaptation for road-object segmentation from a LiDAR point cloud//2019 International Conference on Robotics and Automation (ICRA). Montreal: IEEE: 4376-4382
  60. 60.
    Wu W X, Wang Z Y, Li Z W, Liu W and Li F X. 2020. PointPWC-Net: cost volume on point clouds for (self-)supervised scene flow estimation//16th European Conference on Computer Vision. Glasgow: Springer: 88-107
  61. 61.
    Wu Z R, Song S R, Khosla A, Yu F, Zhang L G, Tang X O and Xiao J X. 2015. 3D ShapeNets: a deep representation for volumetric shapes//2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Boston: IEEE: 1912-1920
  62. 62.
    Xu C F, Wu B C, Wang Z N, Zhan W, Vajda P, Keutzer K and Tomizuka M. 2020. Squeezesegv3: Spatially-adaptive convolution for efficient point-cloud segmentation//Computer Vision Springer: 1-19. Springer International Publishing
  63. 63.
    Xu J L, Xiao L and López A M. 2019. Self-supervised domain adaptation for computer vision tasks. IEEE Access, 7: 156694-156706
  64. 64.
    Xu Q G, Zhou Y, Wang W Y, Qi C R and Anguelov D. 2021. SPG: unsupervised domain adaptation for 3D object detection via semantic point generation//2021 IEEE/CVF International Conference on Computer Vision (ICCV). Montreal: IEEE: 15426-15436
  65. 65.
    Yan H L, Ding Y K, Li P H, Wang Q L, Xu Y and Zuo W M. 2017. Mind the class weight bias: weighted maximum mean discrepancy for unsupervised domain adaptation//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu: IEEE: 945-954
  66. 66.
    Yan Y, Mao Y X and Li B. 2018. SECOND: sparsely embedded convolutional detection. Sensors, 18(10): 3337
  67. 67.
    Yang J H, Shi S S, Wang Z, Li H S and Qi X J. 2021a. ST3D: self-training for unsupervised domain adaptation on 3D object detection//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Nashville: IEEE: 10363-10373
  68. 68.
    Yang J H, Shi S S, Wang Z, Li H S and Qi X J. 2023. ST3D++: denoised self-training for unsupervised domain adaptation on 3D object detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(5): 6354-6371
  69. 69.
    Yang S Q, Wang Y X, van de Weijer J, Herranz L and Jui S L. 2021b. Exploiting the intrinsic neighborhood structure for source-free domain adaptation. arXiv: 2110.04202
  70. 70.
    Yew Z J and Lee G H. 2020. RPM-Net: robust point matching using learned features//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE: 11821-11830
  71. 71.
    Yi L, Gong B Q and Funkhouser T. 2021. Complete and label: a domain adaptation approach to semantic segmentation of LiDAR point clouds//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Nashville: IEEE: 15358-15368
  72. 72.
    Yoo H and Jun K. 2023. Point cloud classification by domain adaptation using recycling max pooling and cutting plane identification. Sensors, 23(3): 1177
  73. 73.
    Yoon D, Tang T M and Barfoot T. 2019. Mapless online detection of dynamic objects in 3D lidar//2019 16th Conference on Computer and Robot Vision (CRV). Kingston: IEEE: 113-120
  74. 74.
    Yuan Z M, Wen C L, Cheng M, Su Y F, Liu W Q, Yu S S and Wang C. 2023. Category-level adversaries for outdoor LiDAR point clouds cross-domain semantic segmentation. IEEE Transactions on Intelligent Transportation Systems, 24(2): 1982-1993
  75. 75.
    Zhang H Y, Cisse M, Dauphin Y N and Lopez-Paz D. 2018. Mixup: beyond empirical risk minimization. arXiv: 1710.09412
  76. 76.
    Zhang W C, Li W and Xu D. 2021. SRDAN: scale-aware and range-aware domain adaptation network for cross-dataset 3D object detection//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Nashville: IEEE: 6765-6775
  77. 77.
    Zhao S C, Wang Y Z, Li B, Wu B C, Gao Y, Xu P F, Darrell T and Keutzer K. 2021. EPointDA: an end-to-end simulation-to-real domain adaptation framework for LiDAR point cloud segmentation. arXiv: 2009.03456
  78. 78.
    Zhuang F Z, Qi Z Y, Duan K Y, Xi D B, Zhu Y C, Zhu H S, Xiong H and He Q. 2021. A comprehensive survey on transfer learning. Proceedings of the IEEE, 109(1): 43-76
  79. 79.
    Zou L K, Tang H, Chen K and Jia K. 2021. Geometry-aware self-training for unsupervised domain adaptation on object point clouds//2021 IEEE/CVF International Conference on Computer Vision (ICCV). Montreal: IEEE: 6383-6392
  80. 80.
    Zou Y, Yu Z D, Vijaya Kumar B V K and Wang J S. 2018. Unsupervised domain adaptation for semantic segmentation via class-balanced self-training//15th European Conference on Computer Vision. Munich: Springer: 297-313

Читать полностью

The above content is generated by Large Model Translation. The translated content is for reference only. We do not assume any commercial or legal responsibilty for any consequences arising from the use of our website