Skip to content

A deep learning approach to find the physical damage on the vehicle such as scratch. The model is developed on the top of COCO Mask RCNN model using a custom dataset

Notifications You must be signed in to change notification settings

arihantkamdar/Damage-detection

Repository files navigation

Damage-detection

A deep learning approach to find the physical damage on the vehicle such as scratch. The model is developed on the top of COCO Mask RCNN model using a custom dataset

Download the COCO Mask RCNN model : https://github.com/matterport/Mask_RCNN/releases/download/v2.0/mask_rcnn_coco.h5

The dataset could be cownloaded from the link given in the below https://drive.google.com/drive/folders/18Ay-ZA5KL-sOqn2GcJZQdKZ7C9pWAOur?usp=sharing

Steps to use repository

  1. Download the dataset and the mask RCNN COCO h5 model from the links given
  2. Change the path of dataset and model in code as mentioned
  3. Install pixellib if it is not installled
  4. Run the code on the demo images given

About

A deep learning approach to find the physical damage on the vehicle such as scratch. The model is developed on the top of COCO Mask RCNN model using a custom dataset

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published