This project contains implementation for Simulated Image Blur from Stereoscopic Images. It runs on Python3.10.2 and uses Jupyter-Notebook for its implementation. The following Python3 libraries are used:
- OpenCV2
- MatPlotLib PyPlot
- NumPy
- SciPy
- iPyWidgets
- Math
To run, launch the Jupyter Notebook file. The last cell launches the user interface, in which you may select a Left and Right image relative to the base directory, change the disparity map values, and change the blurring parameters.
The included images are from the Middlebury Stereoscopic image dataset. Sample image paths are
cones/im2.png
or teddy/im2.png
for the left images, and cones/im6.png
or teddy/im6.png
for
the right images.