https://github.com/s-4-m-a-n/sketch-to-photo-converter

deep learning model that can convert portrait sketch to a portrait photograph
https://github.com/s-4-m-a-n/sketch-to-photo-converter

Last synced: 3 months ago
JSON representation

deep learning model that can convert portrait sketch to a portrait photograph

• Host: GitHub
• URL: https://github.com/s-4-m-a-n/sketch-to-photo-converter
• Owner: s-4-m-a-n
• Created: 2022-09-30T12:45:42.000Z (over 2 years ago)
• Default Branch: master
• Last Pushed: 2022-10-01T14:08:23.000Z (over 2 years ago)
• Last Synced: 2025-01-28T10:37:29.641Z (5 months ago)
• Language: Jupyter Notebook
• Size: 563 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# sketch-to-photo-converter

It deep learning model created by implementing conditional GANs i.e pix2pix. It can convert portrait sketch to a portrait photograph.

# How to use

- create a new folder named model

```$ mkdir model```

- download the trained model from -> https://drive.google.com/file/d/1eTgm0eiE3Ggabdqvl-amcrvzDzt52WuP/view?usp=sharing and put it into the model/

# Dataset

- CUHK sketch face dataset -> http://mmlab.ie.cuhk.edu.hk/archive/facesketch.html 

- pix2pix https://machinelearningmastery.com/how-to-develop-a-pix2pix-gan-for-image-to-image-translation/ 

# Input Image

![sketch](https://github.com/s-4-m-a-n/sketch-to-photo-converter/blob/master/test/inputs/9.jpg)

# result

![output](https://github.com/s-4-m-a-n/sketch-to-photo-converter/blob/master/test/outputs/gen_9.png)

# Conclusion

The model is not robust yet and can not give the acceptable result, that is because of lack in dataset.