臺灣博碩士論文加值系統

This study outlines the steps involved in producing the PCRD dataset, which consists of a set of 3D mesh models and CAD models that have internal shape data. The creation of a game engine environment with optimal lighting and camera settings for collecting the original 3D models and filtering 79,439 raw objects from other 3D datasets—some of which still need to be converted to be checked—represents a substantial contribution of this study. The items that were part of the PCRD dataset were manually filtered to exclude those without interior forms, such as storage spaces, cabinets, drawers, and other categories. The PCRD dataset is unique in that it contains internal shape information, which is often missing from other datasets. This makes the PCRD dataset a valuable resource for researchers and developers working in the field of 3D reconstruction. In conclusion, the PCRD dataset provides a valuable resource for researchers and developers working in the field of 3D reconstruction, particularly those interested in objects with internal structures. In total, 725 objects are reconstructed as the final of this research and much more may yet be added to this in further studies.

這篇研究概述了製作PCRD數據集的步驟，該數據集由一組具有內部形狀數據的3D網格模型和CAD模型組成。創建一個具有最佳照明和攝影機設置的遊戲引擎環境，用於收集原始3D模型並從其他3D數據集中過濾出79,439個原始物體，其中一些仍然需要轉換後進行檢查，這是本研究的重要貢獻。PCRD數據集中的項目是手動過濾的，排除了沒有內部形式的項目，如存儲空間、櫥櫃、抽屜等類別。PCRD數據集的獨特之處在於它包含內部形狀信息，而其他數據集通常缺少此類信息。總的來說，PCRD數據集為3D重建領域的研究人員和開發人員提供了寶貴的資源，特別是對於那些對具有內部結構的物體感興趣的人員。本研究最終重建了725個物體，未來的研究中可能還會增加更多。

摘 要 ii
ABSTRACT iii
Acknowledgement iv
Table Of Contents v
List of Table vi
List of Figure vii
1 Introduction 1
1.1 Research Background 1
1.2 Research Questions and Objectives 3
1.3 Research Contribution 4
1.4 Research Limitation 4
2 Literature Review 5
2.1 Dataset 5
2.1.1 ShapeNet 5
2.1.2 3D-FUTURE 6
2.1.3 Herman Miller 6
2.1.4 ABO 6
2.1.5 IKEA 6
2.1.6 Pix3D 6
2.2 Game Engine 7
2.2.1 Unity 7
2.3 Photogrammetry Software 7
2.3.1 Agisoft Metashape (old name: PhotoScan) 8
3 Research Design and Implementation 9
3.1 Setup & Setting 11
3.2 Chosing & Filtering 13
3.3 Virtual Capturing 15
3.4 Reconstructing 3D Model 19
4 Results 22
4.1 Pairing 22
4.2 Grouping and Sorting 22
4.3 Comparison of Original and Reconstructed Objects 24
5 Conclusion 29
Reference 31

Chang, A. X., Funkhouser, T., Guibas, L., Hanrahan, P., Huang, Q., Li, Z., Savarese, S., Savva, M., Song, S., Su, H., Xiao, J., Yi, L., & Yu, F. (2015). ShapeNet: An Information-Rich 3D Model Repository. http://arxiv.org/abs/1512.03012
Collins, J., Goel, S., Deng, K., Luthra, A., Xu, L., Gundogdu, E., Zhang, X., Vicente, T. F. Y., Dideriksen, T., Arora, H., Guillaumin, M., & Malik, J. (2022). ABO: Dataset and Benchmarks for Real-World 3D Object Understanding. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 21094–21104. https://doi.org/10.1109/CVPR52688.2022.02045
Đurić, I., Vasiljević, I., Obradović, M., Stojaković, V., Kićanović, J., & Obradović, R. (2021). Comparative Analysis of Open-Source and Commercial Photogrammetry Software for Cultural Heritage.
Fu, H., Jia, R., Gao, L., Gong, M., Zhao, B., Maybank, S., & Tao, D. (2021). 3D-FUTURE: 3D Furniture Shape with TextURE. International Journal of Computer Vision, 129(12), 3313–3337. https://doi.org/10.1007/s11263-021-01534-z
Hanke, K., & Grussenmeyer, P. (2002). Digital Photogrammetry. Congress of the International Society for Photogrammetry and Remote Sensing.
Herman-Miller. (2018). Herman Miller 3D Models. Https://Www.Hermanmiller.Com/Resources/Models/3d-Models/.
Ilett, D. (2022). Building Quality Shaders for Unity®. In Building Quality Shaders for Unity®. Apress. https://doi.org/10.1007/978-1-4842-8652-4
Kingsland, K. (2020). Comparative analysis of digital photogrammetry software for cultural heritage. Digital Applications in Archaeology and Cultural Heritage, 18. https://doi.org/10.1016/j.daach.2020.e00157
Lehoczky, M., & Abdurakhmonov, Z. (2021). Present software of photogrammetric processing of digital images. E3S Web of Conferences, 227. https://doi.org/10.1051/e3sconf/202122704001
Setiyadi, S., Mukhtar, H., & Cahyadi, W. A. (2021). A Comparative Study of Affordable Photogrammetry Software for Reconstructing 3D Model of a Human Foot. 2021 IEEE 7th International Conference on Smart Instrumentation, Measurement and Applications, ICSIMA 2021, 286–291. https://doi.org/10.1109/ICSIMA50015.2021.9526314
Su, Y., Liu, M., Rambach, J., Pehrson, A., Berg, A., & Stricker, D. (2021). IKEA Object State Dataset: A 6DoF object pose estimation dataset and benchmark for multi-state assembly objects. http://arxiv.org/abs/2111.08614
Sun, X., Wu, J., Zhang, X., Zhang, Z., Zhang, C., Xue, T., Tenenbaum, J. B., & Freeman, W. T. (2018). Pix3D: Dataset and Methods for Single-Image 3D Shape Modeling. http://arxiv.org/abs/1804.04610
Vîlceanu, C.-B., Pescari, S., Herban, S., & Budău, L. (2022). COMPARATIVE STUDY OF 3D MODELING BY SHORT-RANGE PHOTOGRAMMETRY. In Transactions on HYDROTEHNICS (Vol. 67, Issue 81).
Xiang, Y., Kim, W., Chen, W., Ji, J., Choy, C., Su, H., Mottaghi, R., Guibas, L., & Savarese, S. (2016). ObjectNet3D: A Large Scale Database for 3D Object Recognition (pp. 160–176). https://doi.org/10.1007/978-3-319-46484-8_10