Code for composing two images together with Python can be found in this (image_composition/BasicImageComposition.ipynb). The example includes sample images of lemons and a few backgrounds to compose them on top of.
Find the full tutorial here: Composing Images with Python for Synthetic Datasets
-
Code for training and running inference with the Matterport implementation of Mask R-CNN can be found in this (mask_rcnn/MaskRCNN_TrainAndInference.ipynb).
-
Please download and keep the mask-rcnn code from this link(https://github.com/ahmedfgad/Mask-RCNN-TF2) and keep inside the folder Mask-RCNN-TF2/
-
The example uses a cig-butt dataset that needs to be downloaded separately from here (https://www.immersivelimit.com/datasets/cigarette-butts) a sample for the same can be found in the datasets folder along with the json.
Coco Annotations explanation can be found here - https://cocodataset.org/#format-data Please also follow along this video tutorial to know more about the annotation scheme used for the dataset (Youtube - Coco Dataset Format Complete Walkthrough) - https://www.youtube.com/watch?reload=9&v=h6s61a_pqfM&feature=emb_title
Tutorial content from ImmersiveLimit.com