Research Article
An Automated Tool for Creating Clothing Catalog Databases: MyEyes–Fashion
@INPROCEEDINGS{10.1007/978-3-031-67307-8_5, author={Mariana Carvalho and Daniel Rocha and V\^{\i}tor Carvalho}, title={An Automated Tool for Creating Clothing Catalog Databases: MyEyes--Fashion}, proceedings={Design, Learning, and Innovation. 8th EAI International Conference, DLI 2023, Aalborg, Denmark, November 6--7, 2023, Proceedings}, proceedings_a={DLI}, year={2024}, month={8}, keywords={Blind People Garments Image Segmentation Neural Networks Deep Learning}, doi={10.1007/978-3-031-67307-8_5} }- Mariana Carvalho
Daniel Rocha
Vítor Carvalho
Year: 2024
An Automated Tool for Creating Clothing Catalog Databases: MyEyes–Fashion
DLI
Springer
DOI: 10.1007/978-3-031-67307-8_5
Abstract
In the field of object identification and segmentation the development of automated dataset creation tools holds significant importance. Recognizing the necessity for abundant data, these tools serve to lighten the manual load of data organization, amplifying efficiency and enabling researchers to focus on the essence of their analysis. Integrated into a project that aims to improve the quality of life and well-being of blind people, this paper presents the steps of creating a tool capable of facilitating the data augmentation of databases to feed artificial intelligence algorithms in the identification and extraction of its features. Using a computer vision detection system, a tool was developed to identify and extract garments and catalog them in a database. For user interaction it was created an interface that allows the insertion of elements in the database and the extraction of necessary resources for easy dataset creation. In the segmentation and classification algorithm of the garments, it was obtained an accuracy between 75% and 100% in its recognition.
