When converting the YOLOv8 ONNX model to the DeepStream engine, there is a loss of accuracy, especially for larger objects. However, when converting the model outside of DeepStream, from ONNX to Engine, accuracy is not affected. The same problem happens for yolov4 trained on tao Warning logs when a…

Thank you very much. I changed the parameter value of StateEstimator from 1 to 0 and now it is working correctly.

Could you attach the onnx model and the video source you are using? You’d better attach the comparison of the results and the commands you used to converting the model outside of DeepStream. Thanks

The problem is the loss of accuracy in detecting the bbox itself. Below is an example: Example of model detection converted to tensor outside deepstream: [out] Example of model detection converted to tensor by deepstream: [deep] Video of example: 204.2x54.7.mp4.zip (6.9 MB) Model Onnx: yo…

So is the way you run the demo these two times exactly the same, but only the way of generating the engine is different? Could you attach the command to run the demo and the source code of the libnvdsinfer_custom_impl_Yolo.so? [image] FranJose: !yolo export model=yolo_best.pt format=engine img…

I carried out some more tests and I believed that the problem is with the conversion to the .engine carried out within Deepstream, as the same problem happens when I use either a yoloV8.onnx or a yoloV4.etlt. yoloV8 was trained by the Ultralytics package and was generated in .pt, then converted to O…

[image] FranJose: When I use the .pt model, the generated BBOX are in the correct positions, and the same happens when I use the .etlt model within TAO, the BBOX are also correct. Could you please post the parameters you used in the above 2 scenarios? You can also refer to our FAQ to tune so…

In this case, do you want the parameters for when the inference with yolov4 and v8 was done outside of deepstream?

Yes. And you’d better briefly describe the inference steps outside of deepstream.

For inference with yolov8 with the pt model, the Ultralytics package was used. Command used was: !yolo task=detect mode=predict model=best.pt source=pintura2_204.2x54.7.mp4 imgsz=640 name=yolov8n_v8_50 hide_labels=False To predict the outputs in TAO, the following command was used: yolo_v4 infer…

OK. Could you file a topic on the TAO Forum to learn how to use the tao-deloy to do the inference. It can directly infer the engine file, which can further narrow down the scope. Thanks

[image] yuweiw: tly infer the engine file, which can further narrow down the scope. Thanks I’m doing the tests together with FranJose. The first step was to train a yolov4 on TAO 4.0.4, the docker image used was nvcr.io/nvidia/tao/tao-toolkit:4.0.0-tf1.15.5 . The GPU is a Tesla V100-PCIE-16GB…

Loss of precision to onnx converter for engine by deepstream 6.3

Accelerated Computing Intelligent Video Analytics DeepStream SDK

yuweiw July 15, 2024, 2:56am 7

Could you please post the parameters you used in the above 2 scenarios?
You can also refer to our FAQ to tune some parameters.

Topic		Replies	Views
Problems with Onnx model zoo -> trtexec -> DeepStream 6.0 pipeline DeepStream SDK	16	1173	November 29, 2023
Loading Yolov5 model into DeepStream DeepStream SDK	20	2651	August 3, 2023
Classifier result on onnx doesn't match Deepstream result DeepStream SDK tensorrt , tensorflow , nvbugs , onnx	34	3983	July 28, 2021
Detections change in Deepstream 6.2 DeepStream SDK tensorrt , camera , cuda , kernel , gstreamer , yolo , python , deepstream , deepstream61	21	2203	July 31, 2023
Get wrong infer results while testing yolov4 on deepstream 5.0 DeepStream SDK	45	10227	April 14, 2021
Issue with Converting ONNX Model with different dimensions to TensorRT Engine for DeepStream DeepStream SDK deepstream	21	945	May 23, 2025
Generate engine using onnx2trt, The engine was used to call deepstream, but an error was reported DeepStream SDK onnx	4	359	June 11, 2024
Using custom model in deepstream DeepStream SDK jetson-inference , python , deepstream	39	1105	September 10, 2024
The inference for adding yolov5 in the deepstream example sends an error when yolov5 turns engine Metropolis Microservices for Jetson	24	925	June 7, 2024
Error importing model engine in deepstream TensorRT	4	1119	December 12, 2022

Loss of precision to onnx converter for engine by deepstream 6.3

Related topics