Shopee Image Dataset(Thailand)

Citation Author(s):
Yamin
Thwe
Rajamangala University of Technology Thanyaburi
Submitted by:
Yamin Thwe
Last updated:
Sun, 08/14/2022 - 11:20
DOI:
10.21227/jm68-7821
Data Format:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

Object detection via images has advanced quickly over the last few decades, their detection accuracy, categorization, and localization are not consistent. Achieving fast and accurate detection of fashion products in the e-commerce environment is very important for selecting the right category. This is closely related to customer satisfaction and happiness which is a critical aspect. 

Instructions: 

The images for the initial stage of this study were collected using Shopee Data Scraper, a tool for extracting product data from the Shopee website. This study makes use of five categories. Pants, mid-length dress, hoodie, jacket, and mid-length skirt are included. Total 958 images of products were obtained. We gathered images in a variety of resolutions, with and without background noise, as well as images captured by vendors. On the other hand, we gathered e-commerce photos from Google image according to our categories.

Please cite this paper

===============

MDPI and ACS Style

Thwe, Y.; Jongsawat, N.; Tungkasthan, A. A Semi-Supervised Learning Approach for Automatic Detection and Fashion Product Category Prediction with Small Training Dataset Using FC-YOLOv4. Appl. Sci. 2022, 12, 8068. https://doi.org/10.3390/app12168068

 

AMA Style

Thwe Y, Jongsawat N, Tungkasthan A. A Semi-Supervised Learning Approach for Automatic Detection and Fashion Product Category Prediction with Small Training Dataset Using FC-YOLOv4. Applied Sciences. 2022; 12(16):8068. https://doi.org/10.3390/app12168068

Chicago/Turabian Style

Thwe, Yamin, Nipat Jongsawat, and Anucha Tungkasthan. 2022. "A Semi-Supervised Learning Approach for Automatic Detection and Fashion Product Category Prediction with Small Training Dataset Using FC-YOLOv4" Applied Sciences 12, no. 16: 8068. https://doi.org/10.3390/app12168068