Add compile option to use YOLOv8 ONNX models for testing.

- added qmake option yolo_onnx to use normal YOLOv8 ONNX models. This makes possible to test
  gimbals camera inside without real model.
- reduced confidence threshold requirement in AiEngineInferencevOnnxRuntime from 0.5 to 0.2
- make printing prettier with ONNX Runtime
- removed unnecessary cv::Mat::clone()

Type: Improvement
Issue: https://denyspopov.atlassian.net/browse/AZ-39
This commit is contained in:
Tuomas Järvinen
2024-08-18 16:59:04 +03:00
parent 86c0a7d5c4
commit 022e4a1200
4 changed files with 43 additions and 34 deletions
+2 -2
View File
@@ -4,7 +4,7 @@
### How to convert Azaion AI model file to ONNX format
```bash
yolo export model=azaion-2024-06-28.pt dynamic=False format=onnx imgsz=640,640
yolo export model=azaion-2024-08-13.pt dynamic=False format=onnx imgsz=640,640
```
## How to use application locally on a Linux PC.
@@ -74,7 +74,7 @@ qmake6 && make
```bash
cd autopilot/misc/rtsp_ai_player
qmake6 && make
./rtsp_ai_player ~/azaion/models/azaion/azaion-2024-06-28.onnx
./rtsp_ai_player ~/azaion/models/azaion/azaion-2024-08-13.onnx
```
### Compile and run rtsp_ai_player with YOLOv8 medium model and gimbal camera:
```bash