Hello
How are you?
Thanks for contributing to this project.
Could u provide a inference script for high-resolution (ex: 2040x1920) images with your 512x512 model?
If it is impossible, should we train a model with high-resolution images again?
Trying to change eval_size in config from 512 to 2040 should address this problem.
Re-training is not very necessary, but retraining in higher resolution should improve performance.
Hello How are you? Thanks for contributing to this project. Could u provide a inference script for high-resolution (ex: 2040x1920) images with your 512x512 model? If it is impossible, should we train a model with high-resolution images again?