* Using Rayon to accelarate YOLO post-processing

* Refactor YOLO with outputs format

* Optimize `conf * clss` for yolov5 v6 v7

* Add depth-anything-v2

* Update README.md

* Update CHANGELOG.md
This commit is contained in:
Jamjamjon
2024-07-12 19:46:48 +08:00
committed by GitHub
parent 25d9088e2e
commit edc3a8897c
69 changed files with 1563 additions and 1203 deletions

View File

@ -1,21 +0,0 @@
## Quick Start
```shell
cargo run -r --example rtdetr
```
## Donwload or export ONNX Model
- Export
```bash
pip install -U ultralytics
yolo export model=rtdetr-l.pt format=onnx simplify dynamic opset=16
```
- Download
[rtdetr-l-f16 model](https://github.com/jamjamjon/assets/releases/download/v0.0.1/rtdetr-l-f16.onnx)
## Results
![](./demo.png)

Binary file not shown.

Before

Width:  |  Height:  |  Size: 439 KiB

View File

@ -1,22 +0,0 @@
use usls::{coco, models::RTDETR, Annotator, DataLoader, Options};
fn main() -> Result<(), Box<dyn std::error::Error>> {
// build model
let options = Options::default()
.with_model("rtdetr-l-f16.onnx")?
.with_confs(&[0.4, 0.15])
.with_names(&coco::NAMES_80);
let mut model = RTDETR::new(options)?;
// load image
let x = vec![DataLoader::try_read("./assets/bus.jpg")?];
// run
let y = model.run(&x)?;
// annotate
let annotator = Annotator::default().with_saveout("RT-DETR");
annotator.annotate(&x, &y);
Ok(())
}