臺灣博碩士論文加值系統

近年來隨著3D列印與網際網路快速發展，豐富的三維模型資源，使得迫切需要一個三維模型檢索系統。本論文提出一個基於視圖(view-based)特徵之三維模型檢索系統，採用多張二維影像表示一個三維模型。因為二維影像具有高辨別率，基於視圖特徵演算法在近幾年中成為三維模型檢索領域的熱門議題。不同於以往的三維模型檢索系統，本論文最初只需要對資料庫模型做平移與縮放正規化，因為所提方法具有旋轉不變性，旋轉正規化可以被省略。首先在均勻球體座標視點抓取三維模型多張二維影像，並且利用具有旋轉、縮放、平移不變性的傅立葉描述子(Fourier Descriptor)與 Zernike 矩(Zernike Moments)表示二維影像特徵，再以深度學習中，非監督式自編碼器(Autoencoder)提取每個模型精練過的特徵向量，以提升檢索效能。模型相似度計算採用改良式郝斯多夫距離(Modified Hausdorff Distance)，以克服模型旋轉變量。最後使用普林斯頓形狀評判標準(Princeton Shape Benchmark，PSB)資料庫做檢索效能評判，證明本論文較其他方法有較好的檢索效能。

As 3D printer and Internet are rapid expansion. The increasing number of 3D models on the Internet makes the demand on 3D model retrieval systems. In this thesis, a 3D models retrieval system based on view-based feature is proposed. The view-based 3D model retrieval system represents a 3D model by its projected views. It became a popular branch in 3D model analysis in recent years owing to the high discriminative property of 2D views. Different from previous algorithms that perform the normalization for rotation invariance, we only normalize the translation and the scale of the 3D models in the dataset. Because our method is rotation-invariant, the normalization for rotation is unnecessary and can be ignore. First of all, we create 2D projected views rendered from a sphere with the number of viewpoints uniformly distributed on it. And extract Fourier descriptor and Zernike Moments descriptor to get translation, rotation and scaling invariant feature. After that we utilize unsupervised learning method, Autoencoder, to refine a compact feature vector of each models and improve the retrieval performance. The (dis)similarity between two 3D models is measured by a Modified Hausdorff Distance function to conquer model rotation invariant. Exhaustive experimental results on Princeton Shape Benchmark Dataset confirm that our method achieves better retrieval performance than other methods.

教授推薦書. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
論文口試委員審定書. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
中文摘要. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
英文摘要. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
誌謝. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
目錄. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
表目錄. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
圖目錄. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1 介紹. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2 相關研究. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1 基於模型特徵(Model-based) . . . . . . . . . . . . . . . . . . . . . . 13
2.2 基於視圖特徵(View-based) . . . . . . . . . . . . . . . . . . . . . . . 14
2.3 混和型方法(Hybrid method) . . . . . . . . . . . . . . . . . . . . . . 14
3 三維模型檢索系統. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.1 投影視圖擷取與模型正規化. . . . . . . . . . . . . . . . . . . . . . . 16
3.2 特徵提取. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2.1 輪廓點正規化. . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2.2 傅立葉描述子(Fourier descriptors) . . . . . . . . . . . . . . . 20
3.2.3 Zernike 矩. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.3 自編碼器(Autoencoder) . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.4 相似度計算. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4 實驗與討論. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
4.1 資料庫與評估工具. . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
4.2 實驗設定. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.3 比較評估. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.4 討論. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
5 結論. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
參考文獻. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

[1] Y. Gao, M. Wang, D. Tao, R. Ji, and Q. Dai, “3-d object retrieval and recognition with hypergraph analysis,” IEEE Transactions on Image Processing, vol. 21, no. 9, pp. 4290–4303, 2012.
[2] K. Ding and Y.-H. Liu, “Sphere image for 3-d model retrieval,” IEEE Transactions on Multimedia, vol. 16, no. 5, pp. 1369–1376, 2014.
[3] J. W. Tangelder and R. C. Veltkamp, “A survey of content based 3d shape retrieval methods,” Multimedia tools and applications, vol. 39, no. 3, pp. 441–471, 2008.
[4] Y. Gao and Q. Dai, “View-based 3d object retrieval: challenges and approaches,”IEEE MultiMedia, vol. 3, no. 21, pp. 52–57, 2014.
[5] M. Eitz, R. Richter, T. Boubekeur, K. Hildebrand, and M. Alexa, “Sketch-based shape retrieval.,” ACM Trans. Graph., vol. 31, no. 4, pp. 31–1, 2012.
[6] K. Ding and Y. Liu, “A sketch-based 3d model retrieval system,” in Pattern Recognition (ICPR), 2012 21st International Conference on, pp. 601–604, IEEE, 2012.
[7] F. Wang, L. Kang, and Y. Li, “Sketch-based 3d shape retrieval using convolutional neural networks,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1875–1883, 2015.
[8] P. Daras and A. Axenopoulos, “A 3d shape retrieval framework supporting multimodal queries,” International Journal of Computer Vision, vol. 89, no. 2-3, pp. 229–247, 2010.
[9] K. Ding, W. Wang, and Y. Liu, “3d model retrieval using bag-of-view-words,”Multimedia Tools and Applications, vol. 72, no. 3, pp. 2701–2722, 2014.
[10] X. Bai, S. Bai, Z. Zhu, and L. J. Latecki, “3d shape matching via two layer coding,”IEEE transactions on pattern analysis and machine intelligence, vol. 37, no. 12, pp. 2361–2373, 2015.
[11] M. Körtgen, G.-J. Park, M. Novotni, and R. Klein, “3d shape matching with 3d shape contexts,” in The 7th central European seminar on computer graphics, vol. 3, pp. 5–17, 2003.
[12] F. Tombari, S. Salti, and L. Di Stefano, “Performance evaluation of 3d keypoint detectors,” International Journal of Computer Vision, vol. 102, no. 1-3, pp. 198–220, 2013.
[13] Y. Gao, Q. Dai, and N.-Y. Zhang, “3d model comparison using spatial structure circular descriptor,” Pattern Recognition, vol. 43, no. 3, pp. 1142–1151, 2010.
[14] T. Zaharia and F. J. Preteux, “3d-shape-based retrieval within the mpeg-7 framework,” in Photonics West 2001-Electronic Imaging, pp. 133–145, International Society for Optics and Photonics, 2001.
[15] M. Kazhdan, T. Funkhouser, and S. Rusinkiewicz, “Rotation invariant spherical harmonic representation of 3 d shape descriptors,” in Symposium on geometry processing, vol. 6, pp. 156–164, 2003.
[16] A. E. Johnson and M. Hebert, “Using spin images for efficient object recognition in cluttered 3d scenes,” IEEE Transactions on pattern analysis and machine intelligence, vol. 21, no. 5, pp. 433–449, 1999.
[17] J. Xie, Y. Fang, F. Zhu, and E. Wong, “Deepshape: Deep learned shape descriptor for 3d shape matching and retrieval,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1275–1283, 2015.
[18] Y. Fang, J. Xie, G. Dai, M. Wang, F. Zhu, T. Xu, and E. Wong, “3d deep shape descriptor,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2319–2328, 2015.
[19] P. Shilane, P. Min, M. Kazhdan, and T. Funkhouser, “The princeton shape benchmark,” in Shape modeling applications, 2004. Proceedings, pp. 167–178, IEEE, 2004.
[20] D.-Y. Chen, X.-P. Tian, Y.-T. Shen, and M. Ouhyoung, “On visual similarity based 3d model retrieval,” in Computer graphics forum, vol. 22, pp. 223–232, Wiley Online Library, 2003.
[21] Z. Lian, A. Godil, X. Sun, and H. Zhang, “Non-rigid 3d shape retrieval using multidimensional scaling and bag-of-features,” in 2010 IEEE International Conference on Image Processing, pp. 3181–3184, IEEE, 2010.
[22] K. Lu, N. He, J. Xue, J. Dong, and L. Shao, “Learning view-model joint relevance for 3d object retrieval,” IEEE Transactions on Image Processing, vol. 24, no. 5, pp. 1449–1459, 2015.
[23] P. Papadakis, I. Pratikakis, S. Perantonis, and T. Theoharis, “Efficient 3d shape matching and retrieval using a concrete radialized spherical projection representation,” Pattern Recognition, vol. 40, no. 9, pp. 2437–2452, 2007.
[24] C. Carlson, “Websit: How I Made Wine Glasses from Sunflowers.” http://blog.wolfram.com/2011/07/28/how-i-made-wine-glasses-from-sunflowers/, 2011. [Online; accessed 26-October-2016].
[25] D. Zhang and G. Lu, “An integrated approach to shape based image retrieval,”in Proceedings of 5th Asian Conference on Computer Vision (ACCV), Melbourne, Australia, 2002.
[26] P. J. Van Otterloo, A contour-oriented approach to shape analysis. Prentice Hall International (UK) Ltd., 1991.
[27] C. T. Zahn and R. Z. Roskies, “Fourier descriptors for plane closed curves,”IEEE Transactions on computers, vol. 100, no. 3, pp. 269–281, 1972.
[28] D. Zhang, G. Lu, et al., “A comparative study of fourier descriptors for shape representation and retrieval,” in Proc. of 5th Asian Conference on Computer Vision (ACCV), pp. 646–651, Citeseer, 2002.
[29] A. Tahmasbi, F. Saki, and S. B. Shokouhi, “Classification of benign and malignant masses based on zernike moments,” Computers in Biology and Medicine, vol. 41, no. 8, pp. 726–735, 2011.
[30] G. E. Hinton and R. R. Salakhutdinov, “Reducing the dimensionality of data with neural networks,” Science, vol. 313, no. 5786, pp. 504–507, 2006.
[31] “Ir evaluation methods for retrieving highly relevant documents,”
[32] R. Osada, T. Funkhouser, B. Chazelle, and D. Dobkin, “Matching 3d models with shape distributions,” in Shape Modeling and Applications, SMI 2001 International Conference on., pp. 154–166, IEEE, 2001.
[33] B. K. P. Horn, “Extended gaussian images,” Proceedings of the IEEE, vol. 72, no. 12, pp. 1671–1686, 1984.
[34] S. B. Kang and K. Ikeuchi, “Determining 3-d object pose using the complex extended gaussian image,” in Computer Vision and Pattern Recognition, 1991. Proceedings CVPR’91., IEEE Computer Society Conference on, pp. 580–585, IEEE, 1991.
[35] M. Ankerst, G. Kastenmüller, H.-P. Kriegel, T. Seidl, et al., “Nearest neighbor classification in 3d protein databases.,” in ISMB, vol. 99, pp. 34–43, 1999.
[36] D. Saupe and D. V. Vranić, “3d model retrieval with spherical harmonics and moments,” in Joint Pattern Recognition Symposium, pp. 392–397, Springer, 2001.
[37] D. V. Vranic, “An improvement of rotation invariant 3d-shape based on functions on concentric spheres,” in Image Processing, 2003. ICIP 2003. Proceedings. 2003 International Conference on, vol. 3, pp. III–757, IEEE, 2003.