Skip to main content

an opencv based rewriting of the "transforms" in torchvision package

Project description


  1. This is an opencv based rewriting of the "transforms" in torchvision package.

  2. All functions depend on only cv2 and pytorch (PIL-free). As the article says, cv2 is three times faster than PIL.

  3. Most functions in transforms are reimplemented, except that:

    1. ToPILImage(opencv we used :)), Scale and RandomSizedCrop which are deprecated in the original version are ignored.

    2. The affine transform in the original one only has 5 degrees of freedom, I implement an Affine transform with 6 degress of freedom called RandomAffine6(can be found in master/cvtorchvision/cvtransforms/ The original method(RandomAffine) is still retained and reimplemented with opencv.

    3. My rotate function is clockwise, however the original one is anticlockwise.

    4. Adding some new methods which can be found in Support(the bolded ones).

    5. All the outputs of the opencv version are almost the same as the original one's(test in master/cvtorchvision/cvtransforms/ if name == 'main':...).


  • Compose, ToTensor, ToCVImage, Normalize

  • Resize, CenterCrop, Pad

  • Lambda(doesn't work well in multiprocess in wihdows)

  • RandomApply, RandomOrder, RandomChoice, RandomCrop,

  • RandomHorizontalFlip, RandomVerticalFlip, RandomResizedCrop,

  • FiveCrop, TenCrop, LinearTransformation, ColorJitter,

  • RandomRotation, RandomAffine, RandomAffine6, RandomPerspective

  • RandomGaussianNoise", "RandomPoissonNoise", "RandomSPNoise

  • Grayscale, RandomGrayscale

How to use:

  1. git clone .

  2. Add cvtorchvision to your python path.

  3. Add "from cvtorchvision import cvtransforms" in your pythion file.

  4. You can use all functions as the original version, for example:

    transform = cvtransforms.Compose([
             cvtransforms.RandomAffine(degrees=10, translate=(0.1, 0.1), scale=(0.9, 1.1), shear=(-10, 0),
             cvtransforms.Resize(size=(350, 350), interpolation='BILINEAR'),
             cvtransforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),

more details can be found in the examples of official tutorials.


The multiprocessing used in dataloader of pytorch is not friendly with lambda function in Windows as lambda function can't be pickled (

So the Lambda in may not work properly in Windows.


python >=3.5.2

numpy >=1.10 ('@' operator may not be overloaded before this version)



opencv-contrib-python-3.4.2 (test with this version, but any version of opencv3 is ok, I think)


Welcome to point out and help to fix bugs !

Watches, Stars and Forks won鈥檛 be rejected :smile:

Thanks HongChu who helps a lot.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for opencv-torchvision-transforms-yuzhiyang, version 0.0.1
Filename, size File type Python version Upload date Hashes
Filename, size opencv_torchvision_transforms_yuzhiyang-0.0.1-py3-none-any.whl (21.5 kB) File type Wheel Python version py3 Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page