Skip to content

A deep learning approach for buildings segmentation using detectron2 model developed by facebook research group (META)

License

Notifications You must be signed in to change notification settings

Asv53/Buildings-detection-GeneralizedRCNN

Repository files navigation

Buildings detection using "detectron2" model developed by META

Roboflow is a end-to-end computer vision tool, and we used it in our work.

Architecture

  1. This project aims to detect buildings from the satellite images.

  2. The architecture used is called GeneralizedRCNN (Region-Based Convolutional Neural Network).

  3. This architecture conists of three blocks

    Feature Pyramid Network (FPN)

    Region Proposal Network (RPN)

    ROI Heads

Detailed explination of each block is present in "https://medium.com/@hirotoschwert/digging-into-detectron-2-47b2e794fabd".

Dataset

The dataset we considered is 'building-detection Image Dataset' from Roboflow.

Code

code is available in the build-seg.ipynb file.

Adjust the file paths before executing

Results

Results are available in the results folder.

About

A deep learning approach for buildings segmentation using detectron2 model developed by facebook research group (META)

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published