本論文主要是探討服飾和配搭樣式之檢索方法。首先，在服飾的檢索方法上，我們擷取色彩、紋理、形狀的統計特徵，並運用轉換空間的相似度比對進行檢索，得到了不錯的檢索結果。在配搭樣式的檢索上，我們對使用者喜好的配搭服飾，統計其色彩特徵的交互變異矩陣，用來代表使用者的配搭偏好。根據交互變異矩陣，可計算不同類衣著間(例如：上衣配下裝)的配搭度並排序檢索結果。根據對六位受測者的實驗結果顯示，此方法可達到44%的前十名精確率(AP)，以及0.57的平均精確率(MAP)，具有相當的實用價值。此外，我們也參考了Pantone色彩機構建議的風格化配色方法，並提出了色彩分佈特徵的融合、以及色彩的近似比對方法，能夠有效地找到符合專業配色風格的穿搭組合。

This thesis focuses on the clothing retrieval and stylish collocation of clothes. Initially, some statistical features of color, texture and shape are converted to the eigen space, individually, in which similarity comparison can be conducted, and good result of clothing retrieval can be obtained. Further, clothing collocation are investigated using the color feature based on cross covariance method. A set of desired collocation patterns, each containing a top (like shirt) and a bottom (like skirt/pant), are collected from a specific user. The corresponding clothing images can then be used to extract the color features so as to compute the cross covariance matrix between the tops and the bottoms, based on which the similarity between a top and a bottom could be defined to measure the degree of match for that pair based on the user’s preference patterns. With the similarity measure, it is possible to recommend the user the mostly matched bottoms among all for a given top. An experiment of clothing collocation is conducted for six subjects. 44% top-10 precision rate and 0.57 mean average precision (MAP) can be achieved, which shows great potential in garment industry.
In addition, some color styles, such as the sunrise, sunlight and sunset styles proposed by Pantone organization, are often used in professional clothing recommendation. In this thesis, the fusion of color features and the approximate match in color space are further proposed, and the stylish collocations of clothes can be retrieved effectively.

[1] Yue, J., Li, Z., Liu, L., & Fu, Z. (2011). Content-based image retrieval using color and texture fused features. Mathematical And Computer Modelling, 54(3-4), 1121-1127.

[2] Neetu Sharma, S., Paresh Rawat, S., & Jaikaran Singh, S. (2011). Efficient CBIR Using Color Histogram Processing. Signal & Image Processing : An International Journal, 2(1), 94-112.

[3] Kannan, A., Mohan, V., & Anbazhagan, N. (2010). An Effective Method of Image Retrieval using Image Mining Techniques. The International Journal Of Multimedia & Its Applications, 2(4), 17-26.

[4] "HSV coNeetu Sharma, S., Paresh Rawat, S., & Jaikaran Singh, S. (2011). Efficient CBIR Using Color Histogram Processing. Signal & Image Processing : An International Journal, 2(1), 94-112.

[5] Gevers, T., & Smeulders, A. (2000). PicToSeek: combining color and shape invariant features for image retrieval. IEEE Trans. On Image Process., 9(1), 102-119.

[6] 葉文智、林伯慎(2013)。應用於草稿形狀檢索之環形特徵彈性校準方法。國立台灣科技大學碩士論文。台北

[7] Kwitt, R., Meerwald, P., & Uhl, A. (2011). Efficient Texture Image Retrieval Using Copulas in a Bayesian Framework. IEEE Trans. On Image Process., 20(7), 2063-2077.

[8] Howarth, P., & Rüger, S. (2005). Robust texture features for still-image retrieval. IEE Proc., Vis. Image Process., 152(6), 868.

[9] Wang, W., Zhang, D., Zhang, Y., Li, J., & Gu, X. (2011). Robust Spatial Matching for Object Retrieval and Its Parallel Implementation on GPU. IEEE Trans. Multimedia, 13(6), 1308-1318.

[10] Ashley, J., Flickner, M., Hafner, J., Lee, D., Niblack, W., & Petkovic, D. (1995). The query by image content (QBIC) system. IEEE computers, 28:23-32

[11] J. R. Smith and S. F. Chang, “VisualSeek: a fully automatic content based query system,” in Proceedings of the Fourth ACM International Conference on Multimedia, 1996, pp. 87-98.

[12] R.W. Piccard A. Pentland and S. Sclaroff. Photobook: Content-based manipulation of image databases. International Journal of Computer Vision, 18(3):233–254, 1996.

[13] Wang, X., Yu, Y., & Yang, H. (2011). An effective image retrieval scheme using color, texture and shape features. Computer Standards & Interfaces, 33(1), 59-68.

[14] Lin, C., Chen, R., & Chan, Y. (2009). A smart content-based image retrieval system based on color and texture feature. Image And Vision Computing, 27(6), 658-665.

[15] Liu, G., Li, Z., Zhang, L., & Xu, Y. (2011). Image retrieval based on micro-structure descriptor. Pattern Recognition, 44(9), 2123-2133.

[16] Xiang Bai, Xingwei Yang, Latecki, L., Wenyu Liu, & Zhuowen Tu,. (2010). Learning Context-Sensitive Shape Similarity by Graph Transduction. IEEE Transactions On Pattern Analysis And Machine Intelligence, 32(5), 861-874.

[17] "Cbir.h1" by Jordill2 - Own work. Licensed under CC BY-SA 3.0 via Wikimedia Commons - https://commons.wikimedia.org/wiki/File:Cbir.h1.gif#/media/File:Cbir.h1.gif

[18] Su, C., Chiu, H., & Hsieh, T. (2014). Content Based Images Retrieval Based on HSV Color Space and GLCM. AMM, 644-650, 4287-4290.

[19] Kinnaree, P., Pattanasethanon, S., Thanaputtiwirot, S., & Boontho, S. (2011). RGB Color Correlation Index for Image Retrieval. Procedia Engineering, 8, 36-41.

[20] "Precisionrecall" by Walber - Own work. Licensed under CC BY-SA 4.0 via Wikimedia Commons – https://commons.wikimedia.org/wiki/File:Precisionrecall.svg#/media/File:Precisionrecall.svg

[21] Jain, A., & Vailaya, A. (1996). Image retrieval using color and shape. Pattern Recognition, 29(8), 1233-1244.

[22] Dai Wei Si, Li, H., & Davis, M. (2004). Visual Design In Dress. Tai pei: Shinning Culture Publishing Co.

[23] 崔京遠&柳香蘭 (2010)。時尚大師教你穿衣哲學 = Fashion & color。台北：漢宇國際

[24] Mcoo視覺編者 (2011)。服飾設計配色寶典。台北：佳魁資訊

[25] Eiseman, L. (2011)。配色大師教你穿出你的魅力色彩：找出你的「識別色」，讓自己更出眾的風格配色術。台北：原點出版社