For a project, I have video footage of a person walking through a museum. He films his surroundings using a smartphone and pans mostly horizontally causing a lot of motion blur. For the project, i need to be able to cut out the paintings from the video and match them with paintings from a database. The motion blur really lowers the quality of the cut-out paintings causing a lot of wrong matches. That's why I am looking for a way to remove the motion blur.
Although I have located several sources, I am struggling to incorporate them into my project. The initial source is composed in C++, and I lack the skills required to translate it to Python. The second source is composed in Python, but it employs a user-defined psf.
- https://docs.opencv.org/4.x/d1/dfd/tutorial_motion_deblur_filter.html
- http://amroamroamro.github.io/mexopencv/opencv/weiner_deconvolution_demo_gui.html
They recommended me using OpenCV but other libraries are also allowed. Is there anyone who has a reliable source or a functional implementation for my problem that they can recommend?
