https://github.com/paperspace/dataaugmentationforobjectdetection

Data Augmentation For Object Detection
https://github.com/paperspace/dataaugmentationforobjectdetection

bounding-box data-augmentation deep-learning imagine-augmentation object-detection opencv

Last synced: 14 days ago
JSON representation

Data Augmentation For Object Detection

• Host: GitHub
• URL: https://github.com/paperspace/dataaugmentationforobjectdetection
• Owner: Paperspace
• License: mit
• Created: 2018-09-10T15:57:22.000Z (about 6 years ago)
• Default Branch: master
• Last Pushed: 2020-04-14T21:45:40.000Z (over 4 years ago)
• Last Synced: 2024-10-15T11:22:14.235Z (28 days ago)
• Topics: bounding-box, data-augmentation, deep-learning, imagine-augmentation, object-detection, opencv
• Language: Jupyter Notebook
• Homepage: https://blog.paperspace.com/data-augmentation-for-bounding-boxes/
• Size: 9.93 MB
• Stars: 1,131
• Watchers: 24
• Forks: 318
• Open Issues: 22
• Metadata Files:
  • Readme: readme.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Data Augmentation For Object Detection

Accompanying code for the [Paperspace tutorial series on adapting data augmentation methods for object detection tasks](https://blog.paperspace.com/data-augmentation-for-bounding-boxes/)

## Dependencies

1. OpenCV 3.x

2. Numpy

3. Matplotlib

We support a variety of data augmentations, like.

### Horizontal Flipping

![Horizontal Flip](Images/hflip.png)

### Scaling

![Scaling](Images/scale_aug.png)

### Translation

![Translation](Images/transl_aug.png)

### Rotation

![Rotation](Images/rotate.png)

### Shearing

![Shearing](Images/shear_box.png)

### Resizing

![Resizing](Images/resize.png)

## Quick Start

A quick start tutorial can be found in the file `quick-start.ipynb` in this repo.

## Documentation

A list of all possible transforms and extensive documentation can be found in by opening `docs/build/html/index.html` in your browser or at this [link.](https://augmentationlib.paperspace.com/)