Dream Lover Simulator

The code was written by Chein Chang, Darren Lee, Jamie Wang and Yoyo Zheng.

Outline

1. Introduction
2. Database
3. Backend

Introduction

Motivation

Initially, We came across a video from a YouTuber demonstrating how artificial intelligence can transform a sketch into a realistic photograph, which I found quite interesting. This inspired me to consider the possibility of using a database of photos to allow people to draw their ideal partner and then compare the resulting image to find the person who most closely resembles it, enabling individuals to visualize what their ideal dream partner might look like.

Demonstrations

It is an IOS app, that you can draw a face portrait, then It will find the most similar peoples.

More Detailed

Database

Using Mysql 8.0, WorkBench

1. Unzip mix.7z
2. Create database name "lover_generator"
3. Import mix into db
4. Select gender Male and Female into View
   1. CREATE VIEW `male` AS SELECT * FROM mix WHERE gender = "Male"
   2. CREATE VIEW `female` AS SELECT * FROM mix WHERE gender = "Female"
5. Unzip images data set into E drive
   • both images1024x1024.7z and resized_images1024x1024.7z

Backend

Create Virtual Enviroment

conda create -n fastapi_env python=3.8.0
conda activate fastapi_env

Install dependcy

pip install -r requirment

Start Server

uvicorn fastapi_server:app --reload --host 0.0.0.0

All resource(privated)

Meeting Record