mcnugget

2d images to 3d models, no relation to mcnuggets

setup

python3 -m venv venv
. devenv.sh
pip install -r requirements.txt
cd mcnugget/mcnugget-app/ && npm i

development

backend

python -m mcnugget

frontend

cd mcnugget/mcnugget-app/ && npm start

research

https://colmap.github.io/cli.html#cli https://hub.docker.com/r/freakthemighty/openmvs/~/dockerfile/ https://github.com/rennu/dpg https://grpc.io/docs/quickstart/python.html https://stackoverflow.com/questions/13942069/flask-stream-output-of-function-to-browser

testing

building

docker build --rm -f Dockerfile -t mcnugget:latest .

running in docker

docker run --rm -p 5000:5000 -v `pwd`:/mcnugget -it mcnugget:latest
cd /mcnugget && python3 -m mcnugget

colmap automatic_reconstructor --use_gpu 0 --dense 0 --num_threads 4 --workspace_path /test/tea --image_path /test/tea/images
colmap model_converter --input_path /test/tea/sparse/0 --output_path /test/tea/model.nvm --output_type NVM
/usr/local/bin/OpenMVS/InterfaceVisualSFM model.nvm
/usr/local/bin/OpenMVS/DensifyPointCloud model.mvs
/usr/local/bin/OpenMVS/ReconstructMesh model_dense.mvs
/usr/local/bin/OpenMVS/RefineMesh --resolution-level 1 model_dense_mesh.mvs
/usr/local/bin/OpenMVS/TextureMesh --export-type obj -o model.obj model_dense_mesh_refine.mvs

license

GPLV3