A python library to perform patch augmentation
Project description
patchmentation
Patchmentation is a python library to perform patch augmentation, a data augmentation technique for object detection, that allows for the synthesis of new images by combining objects from one or more source images into a background image.
The process of patch augmentation involves extracting objects of interest from the source images, transforming them, and then pasting them onto the background image to create a composite image, therefore increasing diversity at the object level. The resulting dataset offers a greater variety of object combinations within a single image, making it more robust and accurate when training object detection models.
Installation
The easiest way to install patchmentation is through pip.
pip install patchmentation
Note: Some functionality of patchmentation might not be working on non-Linux systems.
External Links
-
GitHub Repository: https://github.com/Xu-Justin/patchmentation
-
Docs: TBA
-
Research Paper: TBA
-
Benchmarking Results: https://github.com/Xu-Justin/patchmentation-yolov5
This project was developed as part of thesis project, Computer Science, BINUS University.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distribution
Hashes for patchmentation-0.1.11-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 9068533d7a631491197d096ce614ca3c39563fb6a1c8537bf0252ea75cf4a2cf |
|
MD5 | 57dfd9e7af95f9b139f20e3ccf9dd50b |
|
BLAKE2b-256 | dd909851dd8af0b1425d61094dd4ce1a9de04fe4d9a0542d11cbc234f0bfd467 |