Mohammed Innat

@atuleu yes, like to add code example in here https://github.com/keras-team/keras-cv/tree/master/examples/layers/preprocessing for reference.

Hello @atuleu, Any update to this PR?

There is already an active PR for eff-v1 and eff-v2. https://github.com/keras-team/keras-cv/pull/570 https://github.com/keras-team/keras-cv/pull/587

@qlzh727 This PR can be reviewed now. Though I will add some more test cases in the `dice_test.py`, feel free to provide some initial comments. Thanks.

@LukeWood I may not able to work on it due to tight schedule. If it's prioritized, someone may need to take it from here. ( Same goes to [Jaccard PR](https://github.com/keras-team/keras-cv/pull/449)....

> Awesome, thanks! I'll be working on these in the upcoming days then. What's left is the documentation, test cases and implementing the suggestions from the comments, right? Sort of.

@kartik4949 there is some interesting post-processing method for bbox. We can make a to-do list of them and timely include them in keras_cv. Here is the ongoing discussion regarding this....

@old-school-kid That would be cool. But I think according to keras design, it might want to include a separate class for each.. (SE block in currently in progress https://github.com/keras-team/keras-cv/pull/505).

@atuleu > I think bounding box augmentation is erroneous for shear and rotation transform. This will lead to larger bounding boxes than the bounding box covering the sheared/rotated object. Shouldn't...

@LukeWood I used `albumentation` for keypoint augmentation and if I remember correctly I didn't have to normalize keypoint to the image size. The keypoint are usually comes in `xy` pairs....