Are dynamic input shapes a problem for Yolov8 models loaded with readNetFromONNX?

C++
2

dynamic input (or layer) shape does not work with opencv. Dynamic means at inference time.
But with onnx model if input is dynamic you can set input shape :

  1. simplify model
    onnxsim a.onnx a.sim.onnx
  2. use netron to know input and name shape

image
image489×767 14.4 KB

  1. fixed input shape using
python -m onnxruntime.tools.make_dynamic_shape_fixed --dim_param batch --dim_value 1 a.sim.onnx a1.onnx
python -m onnxruntime.tools.make_dynamic_shape_fixed --dim_param width --dim_value 640 a1.onnx a2.onnx
python -m onnxruntime.tools.make_dynamic_shape_fixed --dim_param height --dim_value 480 a2.onnx a3.onnx

Sometimes data input layer are dynamic then post an issue

1 Like