triple-Mu / YOLOv8-TensorRT

YOLOv8 using TensorRT accelerate !
MIT License
1.33k stars 229 forks source link

TypeError: Argument 'thickness' is required to be an integer from infer-det.py #102

Closed sirius2beta closed 1 year ago

sirius2beta commented 1 year ago

Hello, I was running on Nvidia jetson xavier nx with jetpack 5.1.1 installed, I successfully exported Yolov8.onnx and .engine file and ran infer-det.py with good result. However, after I rebuild and installed opencv 4.7.0 with gstreamer support yesterday, I got " TypeError: Argument 'thickness' is required to be an integer " message. I tried to change bbox[:2] to tuple(bbox[:2]), and no more error message was noted. What seems to be the problem? 截圖 2023-05-27 上午9 33 24 截圖 2023-05-27 上午9 41 19

triple-Mu commented 1 year ago

I can not reproduct your error. It seems like you opencv api bugs. Suggest you installing opencv by pip install opencv-python

triple-Mu commented 1 year ago

closed, feel free to reopen it if you have further questions. Thanks!