Azure
FEAT Image Layering Converter
Is your feature request related to a problem? Please describe.
When building testing datasets, we want to be able to create many variations of a specific scenario. For example, we may want to create many samples images layered on top of a base image from a dataset like this one.
Describe the solution you'd like
A converter that takes in a base image, and a secondary image to embed within the main image. (e.g. a photo of hands holding a locket as the base photo, and the secondary image is a CAPTCHA placed over the open locket).
The converter would also likely take a parameter that specifies desired position (e.g. X/Y coordinates of where to put the image on the base image).
@u7702792 of course! Please go ahead and reach out if you have any questions!
@u7702792 of course! Please go ahead and reach out if you have any questions!
Hi, @romanlutz thanks for replying me. I would like to confirm the following problems:
- I think this convert should be inside the prompt_converter folder
- I feel like this image overlaying converter might be similar to an existed converter named add_image_text_converter.py, so shall I follow the coding style/standard like it?
- When I finish this image overlaying converter, should I update the init.py file at the same time while creating the pull request?
Please feel free to point out my mistakes, thanks!
Absolutely! You can also add an example to a notebook if there is one for the other converter.
Absolutely! You can also add an example to a notebook if there is one for the other converter.
Hi, @romanlutz @nina-msft I have created a PR, but I did not fully test my code. It will be grateful if anyone can help me on testing, thanks in advance.