KO
|
EN
gitlite — search
Search
#python
#java
#python3
#arduino
#golang
#machine-learning
#rust
#html
#flask
#javascript
#seismology
#nodejs
nightimaging23
★ 22
Open GitHub ↗
No description available.
Download README (.md)
Explore Similar Repositories
RustOpenAIAPIs
:
No description available.
non-paralell-
:
No description available.
sbase-practice
:
No description available.
ecosystem
:
No description available.
go-fx-test
:
Testing Uber's fx dependency injection system for Go
// repository documentation
Was this content helpful?
★ 0
(0 ratings)
Select Rating:
★
★
★
★
★
Submit Feedback
Recent Feedback
×
Download README
Do you want to download the
README.md
file for
nightimaging23
?
Download (.md)
# Night Photography Rendering Challenge Our challenge is to develop a procedure for creating realistic night scene photos without actual ground-truth images. This task is important for surveillance, security, and art. Submissions will be evaluated by mean opinion scores and professional photographers. This repo contains the source code of [Night Photography Rendering Challenge 2023](https://nightimaging.org/). ## Installation and requirements To work with code it is recommended to use **Python 3.9+**. The required packages are listed in `requirements.txt` and can be installed by calling: ```bash pip install -r requirements.txt ``` ## raw_prc_pipeline Module `raw_prc_pipeline` contains the source code of various methods and functions that can be used for processing raw images, including: - Parsing metadata (see `raw_prc_pipeline/exif_*.py` files) - Demosaicing, white_balancing (Gray World, White Patch, Shades of Gray, [Improved White Patch](https://ieeexplore.ieee.org/document/7025121)), tone mapping and other methods (see `raw_prc_pipeline/pipeline_utils.py`) - Implemtations of demo classes of raw image processing pipeline and image processing pipeline executor (see `raw_prc_pipeline/pipeline.py`) ## data Directory `data` conatins example of challenge data: - PNG file with raw image data (see `data/IMG_2304.png`) and - Corresponding JSON file (see `data/IMG_2304.json`) with necessary metadata including: `black_level`, `white_level`, `cfa_pattern`, `color_matrix_*` and etc. Metadata was extracted using `raw_prc_pipeline` module. ## demo Directory `demo` contains demonstration script for processing PNG raw images with JSON metadata using implemented classes and finctions from `raw_prc_pipeline`. To process PNG images with corresponding metadata from `data` directory call the following command: ```bash python -m demo.process_pngs -p data -ie gw -tm Flash ``` To see other arguments of the script call `python -m demo.process_pngs -h` from the root directory. Also you can use more reproducible way via Docker: ```bash sudo docker build -t nightimaging . sudo docker run --rm -u $(id -u):$(id -g) -v $(pwd)/data:/data nightimaging ./run.sh ``` Also the visualizations of different stages of implemented demo raw image processing pipeline can be found in the `demo/process_img.ipynb` file, which one can [](https://colab.research.google.com/github/createcolor/nightimaging/blob/main/demo/process_img.ipynb).
