mahdi-darvish/Instance-Segmentation-via-Training-Mask-RCNN-on-Custom-Dataset

Combining Google Open Images with COCO-dataset weights and training a Mask R-CNN model to accurately create a instance mask for pumpkins ;)

24
/ 100
Experimental

This project helps you accurately identify and outline individual objects within an image, creating a precise pixel-level mask for each instance rather than just a bounding box. It takes a collection of images and associated annotations (like those from Google Open Images) and outputs a trained model capable of drawing detailed masks around specific objects. This is ideal for anyone working with visual data who needs to distinguish between multiple occurrences of the same object in an image.

No commits in the last 6 months.

Use this if you need to precisely locate and differentiate individual objects within images by creating detailed, pixel-level outlines for each one.

Not ideal if you only need to draw a simple box around objects or if your objects frequently overlap significantly, as it may struggle with highly clumped instances.

image-analysis object-detection visual-inspection agricultural-tech computer-vision-research
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 8 / 25
Community 10 / 25

How are scores calculated?

Stars

15

Forks

2

Language

Jupyter Notebook

License

Last pushed

Mar 28, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/mahdi-darvish/Instance-Segmentation-via-Training-Mask-RCNN-on-Custom-Dataset"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.