A drop-in replacement for Torchvision Transforms using OpenCV
Project description
opencv_transforms
This repository is intended as a faster drop-in replacement for Pytorch's Torchvision augmentations. This repo uses OpenCV for fast image augmentation for PyTorch computer vision pipelines. I wrote this code because the Pillow-based Torchvision transforms was starving my GPU due to slow image augmentation.
Requirements
- A working installation of OpenCV. Tested with OpenCV version 3.4.1, 4.1.0
- Tested on Windows 10 and Ubuntu 18.04. There is evidence that OpenCV doesn't work well with multithreading on Linux / MacOS, for example
num_workers >0
in a pytorchDataLoader
. I haven't run into this issue yet.
Installation
opencv_transforms is now a pip package! Simply use
pip install opencv_transforms
Usage
Breaking change! Please note the import syntax!
from opencv_transforms import transforms
- From here, almost everything should work exactly as the original
transforms
.
Example: Image resizing
import numpy as np
image = np.random.randint(low=0, high=255, size=(1024, 2048, 3))
resize = transforms.Resize(size=(256,256))
image = resize(image)
Should be 1.5 to 10 times faster than PIL. See benchmarks
Performance
- Most transformations are between 1.5X and ~4X faster in OpenCV. Large image resizes are up to 10 times faster in OpenCV.
- To reproduce the following benchmarks, download the Cityscapes dataset.
- An example benchmarking file can be found in the notebook bencharming_v2.ipynb I wrapped the Cityscapes default directories with a HDF5 file for even faster reading.
The changes start to add up when you compose multiple transformations together.
TODO
- Initial commit with all currently implemented torchvision transforms
- Cityscapes benchmarks
- Make the
resample
flag onRandomRotation
,RandomAffine
actually do something - Speed up augmentation in saturation and hue. Currently, fastest way is to convert to a PIL image, perform same augmentation as Torchvision, then convert back to np.ndarray
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 Distribution
opencv_transforms-0.0.3.post1.tar.gz
(917.3 kB
view hashes)
Built Distribution
Close
Hashes for opencv_transforms-0.0.3.post1.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 0f8952a6f73ca719fa0462dd6af012894d437637d5abe6f8c247f52dcd383a38 |
|
MD5 | 0d19eb2f62c88cc2f2d16f53c0af0e6f |
|
BLAKE2b-256 | 4b2ec71d3c8f1880f2a6777006b5b3b84226c77e9676199977aec3ceee7b3683 |
Close
Hashes for opencv_transforms-0.0.3.post1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 75a34364c21c5d1c21d72451cb29f7abf003cc8f6d85dc05a34b832fce04cb98 |
|
MD5 | df857efccdeb1ad05a04298b4faf6229 |
|
BLAKE2b-256 | 4b2c76c609c1d5087b70dc820bc6e3cd640e95aeb540f727c83263d4278c11fd |