kingardor/yolov8-deepstream-6-1

YOLOv8 by Ultralytics in DeepStream 6.1

NVIDIA DeepStream with Ultralytics YOLOv8

Run the latest object-detection SOTA model YOLOv8 with DeepStream!

Repository setup

This is a straightforward step, however, if you are new to git, I recommend glancing threw the steps.

First, install git

sudo apt install git

Next, clone the repository

https://github.com/kn1ghtf1re/yolov8-deepstream-6-1.git

Download the model files

You can run the download-models.sh script to download onnx yolov8 models

cd models
chmod 0777 download-models.sh
./download-models.sh

Docker image build

This repository is tested and used with docker. For docker setup with nvidia, click here.

1. Build the image

cd docker/dgpu

docker build . -t ds-6.1:custom-1

2. Start a container

xhost +

export DISPLAY=:0

docker run -it --rm --net=host --runtime nvidia -e DISPLAY=$DISPLAY -v /tmp/.X11-unix/:/tmp/.X11-unix -v /home/kn1ght/Projects/yolov8-deepstream-6-1:/app ds-6.1:custom-1

Running the Application

1. Build the application

make clean && make -j$(nproc)

2. Start the application

Next, create a file called inputsources.txt and paste the path of videos or rtsp url.

file:///home/astr1x/Videos/sample.mp4
rtsp://admin:admin%[email protected]:554/stream

Now, run the application by running the following command:

# For barebones
./ds-yolov8

# For system76 hybrid graphics
__NV_PRIME_RENDER_OFFLOAD=1 ./ds-yolov8

Citations

• ultralytics/ultralytics

yolov8-deepstream-6-1's Issues

Dependency error in Jetson AGX Xavier.

Hi! thank you very much for your input, I would like to try it but I am having problems with dependencies.

I had to install libboost-filesystem-dev. And now I get another dependency error when compiling.

/usr/bin/ld: can't find -lcurl
/usr/bin/ld: can't find -lgnutls
/usr/bin/ld: can't find -lssl
/usr/bin/ld: unable to find -lcrypto
/usr/bin/ld: unable to find -lboost_program_options
/usr/bin/ld: unable to find -lboost_date_time
/usr/bin/ld: unable to find -lboost_context
/usr/bin/ld: unable to find -lboost_coroutine
/usr/bin/ld: unable to find -lboost_chrono
/usr/bin/ld: unable to find -lboost_log
/usr/bin/ld: unable to find -lboost_thread
/usr/bin/ld: unable to find -lboost_log_setup
/usr/bin/ld: unable to find -lboost_regex
/usr/bin/ld: unable to find -lboost_atomic

Could you help me, I want to run it on a NVIDIA Jetson, best regards.

Custom trained model support :)

Hey @kn1ghtf1re, I was curious if you could share the process of creating the ONNX file with custom pytorch checkpoints so that I can use it for custom models. Thanks!!

How do you include nms into the model?

Hi,

Could you please share how you include NMS in the model?

Thanks.

Does this support instance segmentation?

In the readme we can see a gif of an instance segmentation model but I don't see how to run a deepstream yolov8-seg pipeline with this repo. Is is really supported?