OpenCV DNN Semantic Segmention doesn't produce expected result with Deeplabv3.onnx Model

C++
2

can you explain a bit more ?

is there any pytorch code for this model to compare ? there is

opencv version ? example image / output ?

btw, opencv’s mean values are in [0…255], not in [0…1] !
(they’re subtracted before the scaling)
meaning, you have to use

        blobFromImage(frame, blob, scale, Size(inpWidth, inpHeight), mean * 255, swapRB, false);

in the same way, you have to multiply your std values by 255 before the division here