Inference YOLOV9 add nms batched by TensorRT (Python backend)

[frk-tt]

give me star if my repo helpful for you!
link: https://github.com/thaitc-hust/yolov9-tensorrt

[WongKinYiu]

added to readme.

[WongKinYiu]

https://github.com/thaitc-hust/yolov9-tensorrt/blob/main/torch2onnx.py#L27

you may need use output = output[1] instead.

[laugh12321]

Hello everyone!

I would like to introduce my open-source project - TensoRT-YOLO, a tool for deploying YOLO Series (Support YOLOv9) with Efficient NMS in TensorRT.

Key Features

• Supports FLOAT32, FLOAT16 ONNX export, and TensorRT inference
• Supports YOLOv5, YOLOv8, YOLOv9, PP-YOLOE and PP-YOLOE+
• Integrates EfficientNMS TensorRT plugin for accelerated post-processing
• Utilizes CUDA kernel functions to accelerate preprocess
• Supports C++ and Python inference

[zahidzqj]

https://github.com/thaitc-hust/yolov9-tensorrt/blob/main/torch2onnx.py#L27

you may need use output = output[1] instead.

I found some differences between the results of Pytorch inference and Tensorrt-NMS inference，but I don't know how it happened