Matlab code for solving the radiosity equation for lighting in computer graphics virtual spaces (folder matlab).
The code is related to the examples I show in the YouTube video https://youtu.be/krIVZvzlxUQ Note that the video language is Finnish but that it has English subtitles.
- Simplest example: empty room.
First run "radiosity_emptyroom_Fcomp.m" in Matlab to create the geometric form factor matrix F, which will be saved to a file in the subfolder ./data/. Then you can run "radiosity_emptyroom_BW.m" to create a black-and-white image of a lit room, and "radiosity_emptyroom_color.m" for a color image. If you want more details (smaller patches), you can make n larger in "radiosity_emptyroom_Fcomp.m". Note that too large n value will cause your computer to run out of memory, so it is a good idea to increase n gradually.
- Another example: room with a dividing wall.
First run "radiosity_wall_Fcomp.m" in Matlab to create the geometric form factor matrix F, which will be saved to a file in the subfolder ./data/. Then you can run "radiosity_wall_color.m" to create a color image. If you want more details (smaller patches), you can make halfn larger in "radiosity_wall_Fcomp.m". Note that too large halfn value will cause your computer to run out of memory, so it is a good idea to increase halfn gradually.
- Third example: room with a (levitating) table.
First run "radiosity_table_Fcomp.m" in Matlab to create the geometric form factor matrix F, which will be saved to a file in the subfolder ./data/. Then you can run "radiosity_table_color.m" to create a color image. If you want more details (smaller patches), you can make halfn larger in "radiosity_wall_Fcomp.m". Note that too large halfn value will cause your computer to run out of memory, so it is a good idea to increase halfn gradually.
Python code for solving the radiosity equation for lighting in computer graphics virtual spaces (folder python).
The code is converted to Python from the aforementioned Matlab code.
Prerequisites: Python3, Matplotlib, SciPy
Installing prerequisites:
- create a Python virtual environment with
python3 -m venv radiosity_env, - activate the Python virtual environment with
source python/radiosity_env/bin/activate, - install Matplotlib
pip3 install matplotlib, - install SciPy
pip3 install scipy, - (optional) patch Matplotlib with PR#22046, for example by adding the changes by hand to the files in
radiosity_env/lib/python*/site-packages/.
Then go to the sources folder with cd python/src for examples.
- Simplest example: empty room.
First run "radiosity_emptyroom.py" with python3 radiosity_emptyroom.py to create the geometric form factor matrix F, which will be saved to a file in the subfolder ../data. Then you can run "radiosity_emptyroom_color.py" with python3 radiosity_emptyroom_color.py to create a color image of a lit room. If you want more details (smaller patches), you can make n larger in "radiosity_emptyroom.py". Note that too large n value will cause your computer to run out of memory, so it is a good idea to increase n gradually.
- Iterative version of the empty room.
Run "radiosity_emptyroom_iterative.py" with python3 radiosity_emptyroom.py to use the iterative method for solving radiosity.