As-Projective-As-Possible (APAP) Image Stitching with Moving DLT (CVPR 2013) - Python Implementation
2020.12.28. ~ 2021.01.03.
Local-Homography warping
This program takes a list of images and performs stitching recursively.
This is a re-implementation with Python.
Official Matlab Code here: https://cs.adelaide.edu.au/~tjchin/apap/
The research paper: https://cs.adelaide.edu.au/~tjchin/apap/files/mdlt.pdf
Zaragoza, Julio, et al. "As-projective-as-possible image stitching with moving DLT." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2013.
python == 3.8.5
numpy == 1.19.2
opencv-python == 4.4.0.46
opencv-contrib-python == 4.4.0.46
pillow == 8.0.1
tqdm == 4.50.2
argparse
You can give various types of options, check out "option.py". The "demo"s are tutorials.
Sequential Stitching:
In addition, if the size of the input image is too small, it is difficult to extract feature points, so that an incorrect homography may be estimated.
thanks.
There is room for optimization in the local warping algorithm by using numpy library. (operation speed)
Dae-Young Song
Undergraduate student, Department of Electronic Engineering, Chungnam National University