Many of us have looked at the awe-inspiring New York City skyline and been curious about each building and its history. We believe that people are unable to fully appreciate the skyline without knowing this information. Our program attempts to identify buildings within images of the New York City skyline taking in only smartphone data and a 3D model. By converting the phone data into the model space, we were able to create a projection of what the buildings are supposed to look like at a given location. Then, using alignment techniques, we were then able to outline the building and provide relevant information. We believe we have set the framework for an innovative application.
Read our final report in "final_report.pdf"
Download the 3d model from here.
Save DA12_3D_Buildings_Merged.gml in the directory.
Install requirements by running pip install -r requirements.txt
Run python main.py to generate 4 test examples.