Image size for Google AutoML Training

[tindoantrong]

Hi, I have some images which are lager than  1024 pixels by 1024 pixels.
Following this document: https://cloud.google.com/vertex-ai/docs/image-data/object-detection/prepare-data
Image size should be 1024 pixels by 1024 pixels suggested maximum. If I resize images, the quality of images is low.

What should I do?
Thank you and
Regards

[anjelab]

As this is the required image size and there is a disclaimer in the documentation: 

For images much larger than 1024 pixels by 1024 pixels some image quality may be lost during Vertex AI's image normalization process.

---

This seems to be the limitation of image training. If you really are adamant to retain the image size more than 1024 x 1024 pixels, you may file a feature request and an engineer working closely in Cloud Vision API can assist you.