Applying self transform to image of a dictionary

1

Hi
i have a dictionary defined in the dataloader
def getitem(self, idx):

    image_name = os.path.join(self.__image_path, self.__image_names[idx])
    image = io.imread(image_name)
    joints = self.__mat_data[:-1, :, idx]
    sample = {'image': image, 'joints': joints}
    if self.transform:
        sample = self.transform(sample)
        img = sample['image']
        label = sample['joints']
        
      
    return sample

i want to apply vertical flip and rotation to the image of this.
How do i write down the self transform
is it self.transform(img)=transforms.compose(transforms.randomverticalflip)
or is there a way to do this outside of dataloader?

2

You can use OpenCV library for transformations like these (cv2.flip, cv2.rotate)

3

You can use Albumentations library. It’s very straight forward to use and supports image with different type of targets.