Please use this identifier to cite or link to this item:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/49921Full metadata record
| ???org.dspace.app.webui.jsptag.ItemTag.dcfield??? | Value | Language |
|---|---|---|
| dc.contributor.advisor | 盧信銘(Hsin-Min Lu) | |
| dc.contributor.author | Fu-Hsien Huang | en |
| dc.contributor.author | 黃富纖 | zh_TW |
| dc.date.accessioned | 2021-06-15T12:26:12Z | - |
| dc.date.available | 2022-08-11 | |
| dc.date.copyright | 2020-08-21 | |
| dc.date.issued | 2020 | |
| dc.date.submitted | 2020-08-11 | |
| dc.identifier.citation | Al-Halah, Z., Stiefelhagen, R., Grauman, K. (2017). Fashion forward: Forecasting visual style in fashion. Paper presented at the Proceedings of the IEEE International Conference on Computer Vision. Everingham, M., Eslami, S. A., Van Gool, L., Williams, C. K., Winn, J., Zisserman, A. (2015). The pascal visual object classes challenge: A retrospective. International journal of computer vision, 111(1), 98-136. Gu, X., Wong, Y., Peng, P., Shou, L., Chen, G., Kankanhalli, M. S. (2017). Understanding fashion trends from street photos via neighbor-constrained embedding learning. Paper presented at the Proceedings of the 25th ACM international conference on Multimedia. Guérin, J., Boots, B. (2018). Improving image clustering with multiple pretrained cnn feature extractors. arXiv preprint arXiv:1807.07760. Hadi Kiapour, M., Han, X., Lazebnik, S., Berg, A. C., Berg, T. L. (2015). Where to buy it: Matching street clothing photos in online shops. Paper presented at the Proceedings of the IEEE international conference on computer vision. Han, J., Luo, P., Wang, X. (2019). Deep self-learning from noisy labels. Paper presented at the Proceedings of the IEEE International Conference on Computer Vision. He, K., Zhang, X., Ren, S., Sun, J. (2016). Deep residual learning for image recognition. Paper presented at the Proceedings of the IEEE conference on computer vision and pattern recognition. Hsiao, W.-L., Grauman, K. (2017). Learning the Latent. Paper presented at the Proceedings of the IEEE International Conference on Computer Vision. Inoue, N., Simo-Serra, E., Yamasaki, T., Ishikawa, H. (2017). Multi-label fashion image classification with minimal human supervision. Paper presented at the Proceedings of the IEEE International Conference on Computer Vision Workshops. Ioffe, S., Szegedy, C. (2015). Batch normalization: Accelerating deep network training by reducing internal covariate shift. arXiv preprint arXiv:1502.03167. Jagadeesh, V., Piramuthu, R., Bhardwaj, A., Di, W., Sundaresan, N. (2014). Large scale visual recommendations from street fashion images. Paper presented at the Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining. Kiapour, M. H., Yamaguchi, K., Berg, A. C., Berg, T. L. (2014). Hipster wars: Discovering elements of fashion styles. Paper presented at the European conference on computer vision. Li, P., Li, Y., Jiang, X., Zhen, X. (2019). Two-stream multi-task network for fashion recognition. Paper presented at the 2019 IEEE International Conference on Image Processing (ICIP). Liu, Z., Luo, P., Qiu, S., Wang, X., Tang, X. (2016). Deepfashion: Powering robust clothes recognition and retrieval with rich annotations. Paper presented at the Proceedings of the IEEE conference on computer vision and pattern recognition. Lo, L., Liu, C.-L., Lin, R.-A., Wu, B., Shuai, H.-H., Cheng, W.-H. (2019). Dressing for Attention: Outfit Based Fashion Popularity Prediction. Paper presented at the 2019 IEEE International Conference on Image Processing (ICIP). Ma, Y., Jia, J., Zhou, S., Fu, J., Liu, Y., Tong, Z. (2017). Towards better understanding the clothing fashion styles: A multimodal deep learning approach. Paper presented at the Thirty-First AAAI Conference on Artificial Intelligence. Mall, U., Matzen, K., Hariharan, B., Snavely, N., Bala, K. (2019). GeoStyle: Discovering Fashion Trends and Events. Paper presented at the Proceedings of the IEEE International Conference on Computer Vision. Matzen, K., Bala, K., Snavely, N. (2017). Streetstyle: Exploring world-wide clothing styles from millions of photos. arXiv preprint arXiv:1706.01869. Niculescu-Mizil, A., Caruana, R. (2005). Predicting good probabilities with supervised learning. Paper presented at the Proceedings of the 22nd international conference on Machine learning. Pedamonti, D. (2018). Comparison of non-linear activation functions for deep neural networks on MNIST classification task. arXiv preprint arXiv:1804.02763. Platt, J. (1999). Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods. Advances in large margin classifiers, 10(3), 61-74. Rachmadi, R. F., Purnama, I. (2015). Vehicle color recognition using convolutional neural network. arXiv preprint arXiv:1510.07391. Reynolds, D. A. (2009). Gaussian Mixture Models. Encyclopedia of biometrics, 741. Simo-Serra, E., Fidler, S., Moreno-Noguer, F., Urtasun, R. (2015). Neuroaesthetics in fashion: Modeling the perception of fashionability. Paper presented at the Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Simo-Serra, E., Ishikawa, H. (2016). Fashion style in 128 floats: Joint ranking and classification using weak data for feature extraction. Paper presented at the Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Simonyan, K., Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556. Sudowe, P., Spitzer, H., Leibe, B. (2015). Person attribute recognition with a jointly-trained holistic cnn model. Paper presented at the Proceedings of the IEEE International Conference on Computer Vision Workshops. Takagi, M., Simo-Serra, E., Iizuka, S., Ishikawa, H. (2017). What makes a style: Experimental analysis of fashion prediction. Paper presented at the Proceedings of the IEEE International Conference on Computer Vision Workshops. Uemura, K., Katsurai, M., Maki, H., Goto, R. (2019). Visualization of fashion style relationship of images with tags. Paper presented at the DEIM Forum. Vasileva, M. I., Plummer, B. A., Dusad, K., Rajpal, S., Kumar, R., Forsyth, D. (2018). Learning type-aware embeddings for fashion compatibility. Paper presented at the Proceedings of the European Conference on Computer Vision (ECCV). Veit, A., Alldrin, N., Chechik, G., Krasin, I., Gupta, A., Belongie, S. (2017). Learning from noisy large-scale datasets with minimal supervision. Paper presented at the Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Vittayakorn, S., Yamaguchi, K., Berg, A. C., Berg, T. L. (2015). Runway to realway: Visual analysis of fashion. Paper presented at the 2015 IEEE Winter Conference on Applications of Computer Vision. Yamaguchi, K., Berg, T. L., Ortiz, L. E. (2014). Chic or social: Visual popularity analysis in online fashion networks. Paper presented at the Proceedings of the 22nd ACM international conference on Multimedia. Yamaguchi, K., Hadi Kiapour, M., Berg, T. L. (2013). Paper doll parsing: Retrieving similar styles to parse clothing items. Paper presented at the Proceedings of the IEEE international conference on computer vision. Yamaguchi, K., Kiapour, M. H., Ortiz, L. E., Berg, T. L. (2012). Parsing clothing in fashion photographs. Paper presented at the 2012 IEEE Conference on Computer Vision and Pattern Recognition. Yang, W., Luo, P., Lin, L. (2014). Clothing co-parsing by joint image segmentation and labeling. Paper presented at the Proceedings of the IEEE conference on computer vision and pattern recognition. Zadrozny, B., Elkan, C. (2002). Transforming classifier scores into accurate multiclass probability estimates. Paper presented at the Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining. Zaeemzadeh, A., Rahnavard, N., Shah, M. (2020). Norm-preservation: Why residual networks can become extremely deep? IEEE Transactions on Pattern Analysis and Machine Intelligence. Zhang, X., Jia, J., Gao, K., Zhang, Y., Zhang, D., Li, J., Tian, Q. (2017). Trip outfits advisor: Location-oriented clothing recommendation. IEEE Transactions on Multimedia, 19(11), 2533-2544. Zhao, B., Feng, J., Wu, X., Yan, S. (2017). Memory-augmented attribute manipulation networks for interactive fashion search. Paper presented at the Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Zheng, S., Yang, F., Kiapour, M. H., Piramuthu, R. (2018). Modanet: A large-scale street fashion dataset with polygon annotations. Paper presented at the Proceedings of the 26th ACM international conference on Multimedia. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/49921 | - |
| dc.description.abstract | 現今的社群媒體上充滿了大量服飾穿搭的照片，為了解街頭時尚(Street Fashion)提供了新的管道。為了更了解亞洲街頭時尚，我們建立了Multi-task Fashion Attributes Recognition Network (FARNet)深度學習模型進行多標籤分類(Multi-label Classification)，此模型利用社群網站使用者所提供的含噪標籤(Noisy Labels)及穿搭圖片，辨識圖片中的服裝顏色、類別及花色，並同時處理含噪標籤、將其修正為正確標籤。另外，為提供FARNet模型訓練資料，本研究建立一個全新的資料集，命名為RichWear。此資料集包含322,198張以日本為主的街頭時尚照片，同時含有大量的文字標籤、使用者資訊、服裝品牌，以及使用者加入的Hashtags；資料集中所有圖片的服裝屬性均有含噪標籤，另有一個4,368張圖片的子集，除了含噪標籤外，也加上人工驗證標籤(Human Verified Labels)，供模型訓練使用。實驗結果顯示，我們的方法在測試資料集上的預測表現，顯著優於其他基準線方法。至於街頭流行趨勢分析方面，我們利用FARNet模型預測出來的服裝標籤探索服裝流行趨勢，並且對RichWear資料集的圖片以高斯混合模型（Gaussian Mixture Model, GMM）進行集群分析(Cluster Analysis)，藉此，我們成功找出了日本與其他亞洲地區的街頭流行趨勢，並且發現時尚風格(Fashion Style)會隨著季節而變化，具有時間上的動態性。 | zh_TW |
| dc.description.abstract | There has been an increasing interest in using deep learning and computer vision techniques for fashion recognition. However, most existing methods predict multiple clothing attributes of fashion images individually, rather than simultaneously. In addition, few studies focus the fashion trend analysis on street styles of Asian areas. In this work, we create a new dataset named RichWear, which contains 322,198 street fashion images with massive labels, users’ information, clothing brands, and user-created hashtags. The dataset focuses on street styles of Japan and other Asian areas. In addition to noisy labels, there is also a subset with expert-verified clothing attributes. To improve fashion recognition, we propose a multi-task Fashion Attributes Recognition Network (FARNet) for multi-label classification. Instead of predicting each attribute individually, FARNet simultaneously predicts three types of clothing attributes; meanwhile, it addresses the noisy labels and correct the noisy labels at the same time. Experimental results show that our methods significantly outperform the compared baselines. For street fashion analysis, we use the predicted labels to find clothing trends, and then perform clustering on the images of RichWear to gather visually correlated images. We successfully find street fashion trends as well as discovering style dynamics in Japan and other Asian areas. | en |
| dc.description.provenance | Made available in DSpace on 2021-06-15T12:26:12Z (GMT). No. of bitstreams: 1 U0001-1108202019023500.pdf: 4862209 bytes, checksum: 4ba442aaa5f693d7bf532b3e7ee87e94 (MD5) Previous issue date: 2020 | en |
| dc.description.tableofcontents | 摘要 i Abstract ii Contents iii List of Figures v List of Tables vi Chapter 1 Introduction 1 Chapter 2 Literature Review 5 2.1 Attribute recognition 5 2.2 Image feature extraction 11 2.3 Street fashion analysis 13 2.3.1 Fashion trends discovery and forecast 14 2.3.2 Style discovery and construction 17 2.3.3 Popularity prediction 19 2.4 Street fashion datasets 21 2.5 Research gaps 24 Chapter 3 Dataset 26 3.1 Image collection and cleaning 26 3.2 Label preprocessing and annotation 29 3.3 Datasets combination 31 Chapter 4 Proposed Method 34 4.1 Model architecture 34 4.1.1 The noise correction network 35 4.1.2 The pattern classification network 38 4.2 Loss functions 38 4.2.1 The correction loss 39 4.2.2 The classification loss 40 Chapter 5 Experiments 41 5.1 Evaluation metrics 41 5.2 Baselines 42 5.3 Experimental details 44 5.4 Empirical results 44 5.5 Additional improvements 48 5.5.1 Minority class oversampling 49 5.5.2 Model calibration 50 Chapter 6 Street Fashion Trend Analysis 53 6.1 Trends of colors and patterns 53 6.2 Image clustering 55 6.3 Street style exploration 56 6.3.1 Fashion trends 56 6.3.2 Style dynamics 57 Chapter 7 Conclusions 64 References 66 Appendix 70 A Dataset modification and combination 70 B Minority class oversampling 71 C Model calibration 73 | |
| dc.language.iso | en | |
| dc.subject | 含噪標籤 | zh_TW |
| dc.subject | 深度學習 | zh_TW |
| dc.subject | 多任務學習 | zh_TW |
| dc.subject | 服裝辨識 | zh_TW |
| dc.subject | 多標籤分類 | zh_TW |
| dc.subject | 電腦視覺 | zh_TW |
| dc.subject | 流行趨勢分析 | zh_TW |
| dc.subject | Computer Vision | en |
| dc.subject | Fashion Trend Analysis | en |
| dc.subject | Multi-Task Learning | en |
| dc.subject | Deep Learning | en |
| dc.subject | Fashion Recognition | en |
| dc.subject | Multi-Label Classification | en |
| dc.subject | Noisy Labels | en |
| dc.title | 從街頭看時尚：應用深度學習瞭解流行趨勢 | zh_TW |
| dc.title | From Street Photos to Fashion Trends: A Deep Learning Approach for Fashion Understanding | en |
| dc.type | Thesis | |
| dc.date.schoolyear | 108-2 | |
| dc.description.degree | 碩士 | |
| dc.contributor.coadvisor | 許耀文(Yao-Wen Hsu) | |
| dc.contributor.oralexamcommittee | 蔡志豐(Chih-Fong Tsai) | |
| dc.subject.keyword | 服裝辨識,多標籤分類,含噪標籤,電腦視覺,深度學習,多任務學習,流行趨勢分析, | zh_TW |
| dc.subject.keyword | Fashion Recognition,Multi-Label Classification,Noisy Labels,Computer Vision,Deep Learning,Multi-Task Learning,Fashion Trend Analysis, | en |
| dc.relation.page | 74 | |
| dc.identifier.doi | 10.6342/NTU202003000 | |
| dc.rights.note | 有償授權 | |
| dc.date.accepted | 2020-08-12 | |
| dc.contributor.author-college | 共同教育中心 | zh_TW |
| dc.contributor.author-dept | 統計碩士學位學程 | zh_TW |
| Appears in Collections: | 統計碩士學位學程 | |
Files in This Item:
| File | Size | Format | |
|---|---|---|---|
| U0001-1108202019023500.pdf Restricted Access | 4.75 MB | Adobe PDF |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.