usls/examples/yolo

YOLO-Series

Detection	Instance Segmentation	Pose

Classification	Obb

Head Detection	Fall Detection	Trash Detection

YOLO-World	Face Parsing	FastSAM

Quick Start

# Classify
cargo run -r --example yolo -- --task classify --ver v5  # YOLOv5 
cargo run -r --example yolo -- --task classify --ver v8  # YOLOv8 

# Detect
cargo run -r --example yolo -- --task detect --ver v5  # YOLOv5 
cargo run -r --example yolo -- --task detect --ver v6  # YOLOv6
cargo run -r --example yolo -- --task detect --ver v7  # YOLOv7
cargo run -r --example yolo -- --task detect --ver v8  # YOLOv8
cargo run -r --example yolo -- --task detect --ver v9  # YOLOv9
cargo run -r --example yolo -- --task detect --ver v10 # YOLOv10
cargo run -r --example yolo -- --task detect --ver rtdetr  # YOLOv8-RTDETR
cargo run -r --example yolo -- --task detect --ver v8 --model yolov8s-world-v2-shoes.onnx  # YOLOv8-world

# Pose
cargo run -r --example yolo -- --task pose --ver v8  # YOLOv8-Pose

# Segment
cargo run -r --example yolo -- --task segment --ver v5  # YOLOv5-Segment
cargo run -r --example yolo -- --task segment --ver v8  # YOLOv8-Segment
cargo run -r --example yolo -- --task segment --ver v8 --model FastSAM-s-dyn-f16.onnx  # FastSAM

# Obb
cargo run -r --example yolo -- --task obb --ver v8  # YOLOv8-Obb

other options

--source to specify the input images
--model to specify the ONNX model
--width --height to specify the input resolution
--nc to specify the number of model's classes
--plot to annotate with inference results
--profile to profile
--cuda --trt --coreml --device_id to select device
--half to use float16 when using TensorRT EP

YOLOs configs with Options

Use official YOLO Models

let options = Options::default()
    .with_yolo_version(YOLOVersion::V5)  // YOLOVersion: V5, V6, V7, V8, V9, V10, RTDETR
    .with_yolo_task(YOLOTask::Classify)  // YOLOTask: Classify, Detect, Pose, Segment, Obb
    .with_model("xxxx.onnx")?;

Cutomized your own YOLO model

// This config is for YOLOv8-Segment 
use usls::{AnchorsPosition, BoxType, ClssType, YOLOPreds};

let options = Options::default()
    .with_yolo_preds(
        YOLOPreds {
            bbox: Some(BoxType::Cxcywh),
            clss: ClssType::Clss,
            coefs: Some(true),
            anchors: Some(AnchorsPosition::After),
            ..Default::default()
        }
    )
    .with_model("xxxx.onnx")?;

Other YOLOv8 Solution Models

Model	Weights	Datasets
Face-Landmark Detection	yolov8-face-dyn-f16
Head Detection	yolov8-head-f16
Fall Detection	yolov8-falldown-f16
Trash Detection	yolov8-plastic-bag-f16
FaceParsing	yolov8-face-parsing-dyn	CelebAMask-HQ [Processed YOLO labels][Python Script]

Export ONNX Models

YOLOv5

Here

YOLOv6

Here

YOLOv7

Here

YOLOv8

pip install -U ultralytics

# export onnx model with dynamic shapes
yolo export model=yolov8m.pt format=onnx simplify dynamic
yolo export model=yolov8m-cls.pt format=onnx simplify dynamic
yolo export model=yolov8m-pose.pt format=onnx simplify dynamic
yolo export model=yolov8m-seg.pt format=onnx simplify dynamic
yolo export model=yolov8m-obb.pt format=onnx simplify dynamic

# export onnx model with fixed shapes
yolo export model=yolov8m.pt format=onnx simplify
yolo export model=yolov8m-cls.pt format=onnx simplify
yolo export model=yolov8m-pose.pt format=onnx simplify
yolo export model=yolov8m-seg.pt format=onnx simplify
yolo export model=yolov8m-obb.pt format=onnx simplify

YOLOv9

Here

YOLOv10

Here