Leveraging Weakly Annotated Data for Fashion Image Retrieval and Label Prediction

Charles Corbiere, Hedi Ben-Younes, Alexandre Rame, Charles Ollion
2017 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)  
In this paper, we present a method to learn a visual representation adapted for e-commerce products. Based on weakly supervised learning, our model learns from noisy datasets crawled on e-commerce website catalogs and does not require any manual labeling. We show that our representation can be used for downward classification tasks over clothing categories with different levels of granularity. We also demonstrate that the learnt representation is suitable for image retrieval. We achieve nearly
more » ... tate-of-art results on the DeepFashion In-Shop Clothes Retrieval and Categories Attributes Prediction [12] tasks, without using the provided training set.
doi:10.1109/iccvw.2017.266 dblp:conf/iccvw/CorbiereBRO17 fatcat:bosdwhsoojcgjoyh7rd2gmgmsy