Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

[Feat] 3D Data Augmentation APIs

See original GitHub issue

🚀 Feature

I am going to give some basic 3D data augmentation methods recently.

The first thing is shall we rename all the current APIs into 2D?

import kornia.augmentation as K
aug_2d  = K.RandomHorizontalFlip()
aug_3d = K.RandomHorizontalFlip3D()

or we shall name it as:

import kornia.augmentation as K
aug_2d  = K.RandomHorizontalFlip2D()
aug_3d = K.RandomHorizontalFlip3D()

Also, the functionalities like kornia.geometry.VFlip accepts only BCHW and I do not think it is appropriate to make it accepts BCDHW. Thus, It seems to be redundant but should we have a 3D version like kornia.geometry.VFlip3D and kornia.geometry.VFlip2D. Or should we have another kornia_3d repo?

Issue Analytics

State:
Created 3 years ago
Comments:8 (5 by maintainers)

Top GitHub Comments

2reactions

edgarribacommented, Jun 25, 2020

/cc @wyli in case you are interested

0reactions

edgarribacommented, Oct 19, 2020

already introduced in v0.4.x

Top Results From Across the Web

3D Data-Augmentation Using tf.data and Volumentations-3D ...

So see it's pretty simple to use augmentation using tf.data API with little effort using Volumentations-3D. Hope you will find it useful.

mmpose.apis — MMPose 0.29.0 documentation

Inference 3D pose from 2D pose sequences using a pose lifter model. Parameters. model (nn.Module) – The loaded pose lifter model. pose_results_2d (list[list[ ......

Building powerful image classification models using very little ...

ImageDataGenerator for real-time data augmentation; layer freezing and model fine-tuning ...and more. Our setup: only 2000 training examples ( ...

Tensorflow CNN image augmentation pipeline - Stack Overflow

I'm using the Dataset API creating 2 datasets so that in the trainloop I can calculate the train and test loss and log...

Improving Deep Learning with Generic Data Augmentation

Experimental results indicate that cropping in geometric augmentations significantly increases CNN task performance, and various geometric ...