A Django web app with MaskRCNN model for object detection and segementation. Developed for gsoc 2020 code challenge
Place the weights for MaskRCNN inside detector/saved_models. https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
- Create an virtual env or conda environment.(optional)
- clone the repo and install dependencies listed in requirements.txt by 'pip install -r requirements.txt'
- Copy the weights of the model into object_detector/object_detector/detector/saved_models/
- To start a local webserver run, 'python manage.py runserver' The app can be now accessed via browser on http://localhost:8000/ (by default)