DAN
The project does not use any machine learning that I can see. The write up says it uses PyTorch or Tensorflow for machine learning and optimization, but I cannot see where either library is used. As far as I can tell, no ML library is imported. Originally I consented to this project because you assured me ML would used, but it has not been. This is a problem for an ML course. I wonder why you state in your writeup that you used PyTorch or Tensorflow when it appears you did not.
You have also not used the principles of workflows that we learned. There is no main.R or main.py. The readme is only a couple lines long and gives no real information. I cannot even tell which file I should run first if I wanted to reproduce your code. The methods in the writeup simply say ``We applied machine learning techniques to improve accuracy for simulating the camera during the monoplotting software in photos that are especially challenging to georeference. The code is available in GitHub Repository. In this repository, there are the code, data and README file to accomplish this process.'' But nothing is said about how ML was used.
The result appears obviously flawed. All the control points in the image are in the foreground on the vegetation. Many are at or near the tops of trees. Since trees are not represented in digital elevation maps, it seems highly unlikely that the result obtained even passes basic and superficial sanity tests for correctness.
Only one image was considered. Even putting aside all the other issues I have identified, I'm afraid the workload involved was not enough even for one student's final project, let alone two.
I wish you had spoken to me about this project while it was underway. I might have been able to help you change course or find an alternative project that could have led to more success.
Grade: 15/30
Programming languages: Python and R studio
Data: Historical images from Uintas, Utah.