spacewalk01 / depth-anything-tensorrt

TensorRT implementation of Depth-Anything V1, V2

Home Page:https://depth-anything.github.io/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

How to export onnx with dynamic shape?

HSQ79815 opened this issue · comments

How to export onnx with dynamic shape?

Currently doesn't support dynamic shape.

But you can create your model based on your video resolution as follows:

python export.py --encoder vitb --load_from depth_anything_vitb14.pth --image_shape 3 video_height video_width