Giters

zhyever / PatchFusion

[CVPR 2024] An End-to-End Tile-Based Framework for High-Resolution Monocular Metric Depth Estimation

Home Page:https://zhyever.github.io/patchfusion/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

How to finetune on low resolution images？

Xie-PC opened this issue · comments

commented

Thank you for your good work.
Now I deal with the depth estimation of 640*360 RGB images, and the ground truth is also of the same size. Your work involves cropped patch of high-resolution images (such as 4K). If I want to finetune my usage scene, I will face particularly small problems after crooping patch. If I first upsample the image to 4K, but I am worried about some inaccurate position values in RGB and depth maps due to interpolation, and artifacts on object boundaries, do you have any treatment methods or suggestions for applying your work to such low-resolution images?

commented

Thanks for your interest in our work. It's indeed hard to directly adopt PatchFusion without any modification on low resolution datasets. One possible way would be proposed in #10. I would recommand Marigold based on your purpose.