NVIDIA DeepStream SDK 8.0 / 7.1 / 7.0 / 6.4 / 6.3 / 6.2 / 6.1.1 / 6.1 / 6.0.1 / 6.0 application for YOLO-Seg models
YOLO object detection models and other infos: https://github.com/marcoslucianops/DeepStream-Yolo
Important: Please export the ONNX model with the new export file, generate the TensorRT engine again with the updated files, and use the new config_infer_primary file according to your model
- Supported models
- Instructions
- YOLOv5-Seg usage
- YOLOv7-Seg usage
- YOLOv7-Mask usage
- YOLOv8-Seg usage
- YOLO11-Seg usage
- RF-DETR-Seg usage
- NMS configuration
- Detection threshold configuration
git clone https://github.com/marcoslucianops/DeepStream-Yolo-Seg.git
cd DeepStream-Yolo-Seg
2.1. Set the CUDA_VER according to your DeepStream version
export CUDA_VER=XY.Z
-
x86 platform
DeepStream 8.0 = 12.8 DeepStream 7.1 = 12.6 DeepStream 7.0 / 6.4 = 12.2 DeepStream 6.3 = 12.1 DeepStream 6.2 = 11.8 DeepStream 6.1.1 = 11.7 DeepStream 6.1 = 11.6 DeepStream 6.0.1 / 6.0 = 11.4 -
Jetson platform
DeepStream 8.0 = 13.0 DeepStream 7.1 = 12.6 DeepStream 7.0 / 6.4 = 12.2 DeepStream 6.3 / 6.2 / 6.1.1 / 6.1 = 11.4 DeepStream 6.0.1 / 6.0 = 10.2
2.2. Make the libs
make -C nvdsinfer_custom_impl_Yolo_seg clean && make -C nvdsinfer_custom_impl_Yolo_seg
deepstream-app -c deepstream_app_config.txt
NOTE: The TensorRT engine file may take a very long time to generate (sometimes more than 10 minutes).
For now, the NMS is configured in the ONNX exporter file.
NOTE: Make sure to set cluster-mode=4 in the config_infer file.
The minimum detection confidence threshold is configured in the ONNX exporter file. The pre-cluster-threshold should be >= the value used in the ONNX model.
[class-attrs-all]
pre-cluster-threshold=0.25
My projects: https://www.youtube.com/MarcosLucianoTV