https://github.com/jrieke/shape-detection

🟣 Object detection of abstract shapes with neural networks
https://github.com/jrieke/shape-detection

dataset deep-learning keras machine-learning machine-learning-tutorials neural-networks object-detection object-recognition synthetic-data tutorial

Last synced: 5 days ago
JSON representation

🟣 Object detection of abstract shapes with neural networks

• Host: GitHub
• URL: https://github.com/jrieke/shape-detection
• Owner: jrieke
• License: mit
• Created: 2016-07-02T13:59:49.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2020-10-22T22:40:03.000Z (about 4 years ago)
• Last Synced: 2024-10-23T00:36:15.454Z (14 days ago)
• Topics: dataset, deep-learning, keras, machine-learning, machine-learning-tutorials, neural-networks, object-detection, object-recognition, synthetic-data, tutorial
• Language: Jupyter Notebook
• Homepage:
• Size: 1.12 MB
• Stars: 218
• Watchers: 10
• Forks: 131
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# shape-detection

**Read my [Medium post](https://towardsdatascience.com/object-detection-with-neural-networks-a4e2c46b4491) on this project**

*Note: Some people reached out and told me that they can't replicate some of the results in this repo. I couldn't find the reason, potentially one of the dependencies changed since I did this project a few years ago. Unfortunately I don't have any notes on which exact setup I used back then. If you find anything, please reach out to me :)*

This is an example of object detection with neural networks (implemented with [keras](https://keras.io/)). The training images contain abstract geometric shapes and can be easily bootstraped. 

The code is split up into several Jupyter notebooks. They increase in complexity:

* Detection of rectangles in numpy arrays: [single-rectangle](single-rectangle.ipynb), [two-rectangles](two-rectangles.ipynb), [multiple-rectangles](multiple-rectangles.ipynb)

* Additional classification between rectangles and triangles: [two-rectangles-or-triangles](two-rectangles-or-triangles.ipynb), [multiple-rectangles-or-triangles](multiple-rectangles-or-triangles.ipynb)

* Application to more complex images (using convnets): [color-multiple-shapes](color-multiple-shapes.ipynb)

![](plots/bw-single-rectangle_prediction.png)

**Requirements**: Python 2.7, keras (v1.0.5), theano or tensorflow, numpy, matplotlib, jupyter, pycairo (only for color-multiple-shapes.ipynb)