- export YOLOv8 model into torchscript, see models/yolo.sh or take a look at quick start section Obtain YOLOv8
- download libtorch, see https://pytorch.org/
The following library is tested against stable version libtorch-2.4.0.
In case you want to use opencv, please install it first:
apt install libopencv-dev clang libclang-devvirtualenv ultra
. ultra/bin/activate
pip3 install ultralytics
yolo export model=yolov8n.pt
ls -l yolov8n.torchscriptMore on yolo export: https://docs.ultralytics.com/modes/export/#__tabbed_1_2
Get libtorch from https://pytorch.org/, the current bindings are using stable version of 2.4.0. Please decide beforehand if you want to use cpu or cuda version.
unzip ~/Downloads/libtorch-cxx11-abi-shared-with-deps-2.4.0+cpu.zipsource ./env # environment
cargo build --releaseThe video example needs to have opencv installed and enabled as feature.
source ./env # environment
cargo build --release --examples --features=opencvsource ./env # environment
target/release/examples/yolo-predictsource ./env # environment
target/release/examples/video <some.video.file>This library tries to do things as close as ultralytics python package.
Two options are available: using torch and tensor, or using opencv. In the latter case the opencv feature must be turned on (default is off).
We pad the image exactly as ultralytics does, i.e. the aspect ratio is preserved and the gray (114, 114, 114) coler
is used as padding to keep (640, 640) as input dimension.
We try to use torch and tensors as long as possible. However, in certain cases it might be faster to move the prediction to the cpu. In that case, please initialize yolo as:
let yolo = yolo_v8::YoloV8ObjectDetection::with_model(yolo_v8::YOLOModel::Nano).post_process_on_cpu();