https://github.com/engineeringmadness/nightlight-web
Fully Convolutional pipeline to enhance pitch dark images
https://github.com/engineeringmadness/nightlight-web
Last synced: about 1 month ago
JSON representation
Fully Convolutional pipeline to enhance pitch dark images
- Host: GitHub
- URL: https://github.com/engineeringmadness/nightlight-web
- Owner: engineeringmadness
- Created: 2019-04-28T05:35:28.000Z (about 7 years ago)
- Default Branch: master
- Last Pushed: 2023-04-08T03:44:05.000Z (about 3 years ago)
- Last Synced: 2025-11-20T20:05:26.776Z (7 months ago)
- Language: Python
- Homepage:
- Size: 1.87 MB
- Stars: 1
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Nightlight Web App
Bachelors of Technology Major project. In this project we have implemented a CNN based on the paper Learning to see in the dark.
The CNN uses a U-Net architecture and takes as input dark images and brightens them up so all the details are visible
Steps to run
1. Clone this repo
2. Install Anaconda and Install additional dependencies by running
>conda install tensorflow
>python -m pip install django pillow rawpy imagio exifread
3. Download model files from below link and place them in transform/model/ folder
4. Download static assets and place them in static/assets/ folder
5. Setup database using
>python manage.py migrate
6. Run app using
>python manage.py runserver
7. Goto 127.0.0.1:8000 in browser
Based on paper Learning to See in the Dark - https://arxiv.org/abs/1805.01934