Overview

For the project, a dashboard where the user can get clothing recommendations based on the image that is uploaded in the Dropbox was developed. Streamlit, a popular open-source Python library designed to create web applications for data exploration, visualization, prototyping machine learning models, and more, was used for the development of the application.

Though the input image is limited at the moment, the system can be used for various recommendations of a wide variety with more input data updated. It is also a good method of understand the DenseNet121 and recommendation systems.

Description: Streamlit page

The initial page of the recommendation dashboard includes the following features:

• Title of the dashboard

• Description of the dashboard (How to use the recommendation app)

• Image Dropbox for uploading images

  

When a user drags or selects the image of a specific type of clothing that is close to their preference in the Dropbox and presses the “Generate recommendation” button, the system will give five recommendation images that are similar to the input image.

Below is an example of getting a recommendation for a certain style of shoes.

1. Drag or upload an image of a specific style of shoe that you would like to get a recommendation for.

1. Get various recommendations from the app, and consider these when making a purchase.

제목 없는 동영상 - Clipchamp로 제작 (2).mp4