flystem
/
flystem-usls
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Docs update

This commit is contained in:
jamjamjon 2024-07-01 16:53:36 +08:00
parent 20d2edce6a
commit f6fec32c7b
21 changed files with 227 additions and 64 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
CHANGELOG.md Normal file
View File

@ -0,0 +1,21 @@
## v0.0.4 - 2024-06-30
### Added
- Add X struct to handle input and preprocessing
- Add Ops struct to manage common operations
- Use SIMD (fast_image_resize) to accelerate model pre-processing and post-processing.YOLOv8-seg post-processing (~120ms => ~20ms), Depth-Anything post-processing (~23ms => ~2ms).
### Deprecated
- Mark `Ops::descale_mask()` as deprecated.
### Fixed
### Changed
### Removed
### Refactored
### Others

98
README.md
View File

@ -1,43 +1,39 @@
# usls # usls
[![Static Badge](https://img.shields.io/crates/v/usls.svg?style=for-the-badge&logo=rust)](https://crates.io/crates/usls) [![Static Badge](https://img.shields.io/badge/Documents-usls-blue?style=for-the-badge&logo=docs.rs)](https://docs.rs/usls) [![Static Badge](https://img.shields.io/badge/GitHub-black?style=for-the-badge&logo=github)](https://github.com/jamjamjon/usls) ![Static Badge](https://img.shields.io/crates/d/usls?style=for-the-badge)
A Rust library integrated with **ONNXRuntime**, providing a collection of **Computer Vison** and **Vision-Language** models including [YOLOv5](https://github.com/ultralytics/yolov5), [YOLOv8](https://github.com/ultralytics/ultralytics), [YOLOv9](https://github.com/WongKinYiu/yolov9), [YOLOv10](https://github.com/THU-MIG/yolov10), [RTDETR](https://arxiv.org/abs/2304.08069), [CLIP](https://github.com/openai/CLIP), [DINOv2](https://github.com/facebookresearch/dinov2), [FastSAM](https://github.com/CASIA-IVA-Lab/FastSAM), [YOLO-World](https://github.com/AILab-CVC/YOLO-World), [BLIP](https://arxiv.org/abs/2201.12086), [PaddleOCR](https://github.com/PaddlePaddle/PaddleOCR), [Depth-Anything](https://github.com/LiheYoung/Depth-Anything), [MODNet](https://github.com/ZHKKKe/MODNet) and others. A Rust library integrated with **ONNXRuntime**, providing a collection of **Computer Vison** and **Vision-Language** models including [YOLOv5](https://github.com/ultralytics/yolov5), [YOLOv8](https://github.com/ultralytics/ultralytics), [YOLOv9](https://github.com/WongKinYiu/yolov9), [YOLOv10](https://github.com/THU-MIG/yolov10), [RTDETR](https://arxiv.org/abs/2304.08069), [CLIP](https://github.com/openai/CLIP), [DINOv2](https://github.com/facebookresearch/dinov2), [FastSAM](https://github.com/CASIA-IVA-Lab/FastSAM), [YOLO-World](https://github.com/AILab-CVC/YOLO-World), [BLIP](https://arxiv.org/abs/2201.12086), [PaddleOCR](https://github.com/PaddlePaddle/PaddleOCR), [Depth-Anything](https://github.com/LiheYoung/Depth-Anything), [MODNet](https://github.com/ZHKKKe/MODNet) and others.
## Recently Updated
| YOLOv8-Obb |
| :----------------------------: |
|<img src='examples/yolov8/demo-obb-2.png' width="800px">|
| Depth-Anything | | Depth-Anything |
| :----------------------------: | | :----------------------------: |
|<img src='examples/depth-anything/demo.png' width="800px">| |<img src='examples/depth-anything/demo.png' width="800px">|
| YOLOP-v2 | Text-Detection |
| :----------------------------: | :------------------------------: |
|<img src='examples/yolop/demo.png' width="385px">| <img src='examples/db/demo.png' width="385x"> |
| Portrait Matting | | Portrait Matting |
| :----------------------------: | | :----------------------------: |
|<img src='examples/modnet/demo.png' width="800px">| |<img src='examples/modnet/demo.png' width="800px">|
| YOLOv8-Obb |
| :----------------------------: |
|<img src='examples/yolov8/demo-obb-2.png' width="800px">|
| YOLOP-v2 | Face-Parsing | Text-Detection |
| :----------------------------: | :------------------------------: | :------------------------------: |
|<img src='examples/yolop/demo.png' height="180px">| <img src='examples/face-parsing/demo.png' height="180px"> | <img src='examples/db/demo.png' height="180px"> |
- 2024/06/30: **Accelerate model pre-processing and post-processing using SIMD**. YOLOv8-seg post-processing (~120ms => ~20ms), Depth-Anything post-processing (~23ms => ~2ms).
## Supported Models ## Supported Models
| Model | Task / Type | Example | CUDA<br />f32 | CUDA<br />f16 | TensorRT<br />f32 | TensorRT<br />f16 | | Model | Task / Type | Example | CUDA<br />f32 | CUDA<br />f16 | TensorRT<br />f32 | TensorRT<br />f16 |
| :---------------------------------------------------------------: | :-------------------------: | :----------------------: | :-----------: | :-----------: | :------------------------: | :-----------------------: | | :---------------------------------------------------------------: | :-------------------------: | :----------------------: | :-----------: | :-----------: | :------------------------: | :-----------------------: |
| [YOLOv5](https://github.com/ultralytics/yolov5) | Object Detection<br />Instance Segmentation<br />Classification | [demo](examples/yolov5) | ✅ | ✅ | ✅ | ✅ | | [YOLOv5](https://github.com/ultralytics/yolov5) | Classification<br />Object Detection<br />Instance Segmentation | [demo](examples/yolov5) | ✅ | ✅ | ✅ | ✅ |
| [YOLOv8-obb](https://github.com/ultralytics/ultralytics) | Object Detection<br />Instance Segmentation<br />Classification<br />Oriented Object Detection<br />Keypoint Detection | [demo](examples/yolov8) | ✅ | ✅ | ✅ | ✅ | | [YOLOv8](https://github.com/ultralytics/ultralytics) | Object Detection<br />Instance Segmentation<br />Classification<br />Oriented Object Detection<br />Keypoint Detection | [demo](examples/yolov8) | ✅ | ✅ | ✅ | ✅ |
| [YOLOv9](https://github.com/WongKinYiu/yolov9) | Object Detection | [demo](examples/yolov9) | ✅ | ✅ | ✅ | ✅ | | [YOLOv9](https://github.com/WongKinYiu/yolov9) | Object Detection | [demo](examples/yolov9) | ✅ | ✅ | ✅ | ✅ |
| [YOLOv10](https://github.com/THU-MIG/yolov10) | Object Detection | [demo](examples/yolov10) | ✅ | ✅ | ✅ | ✅ | | [YOLOv10](https://github.com/THU-MIG/yolov10) | Object Detection | [demo](examples/yolov10) | ✅ | ✅ | ✅ | ✅ |
| [RT-DETR](https://arxiv.org/abs/2304.08069) | Object Detection | [demo](examples/rtdetr) | ✅ | ✅ | ✅ | ✅ | | [RTDETR](https://arxiv.org/abs/2304.08069) | Object Detection | [demo](examples/rtdetr) | ✅ | ✅ | ✅ | ✅ |
| [FastSAM](https://github.com/CASIA-IVA-Lab/FastSAM) | Instance Segmentation | [demo](examples/fastsam) | ✅ | ✅ | ✅ | ✅ | | [FastSAM](https://github.com/CASIA-IVA-Lab/FastSAM) | Instance Segmentation | [demo](examples/fastsam) | ✅ | ✅ | ✅ | ✅ |
| [YOLO-World](https://github.com/AILab-CVC/YOLO-World) | Object Detection | [demo](examples/yolo-world) | ✅ | ✅ | ✅ | ✅ | | [YOLO-World](https://github.com/AILab-CVC/YOLO-World) | Object Detection | [demo](examples/yolo-world) | ✅ | ✅ | ✅ | ✅ |
| [DINOv2](https://github.com/facebookresearch/dinov2) | Vision-Self-Supervised | [demo](examples/dinov2) | ✅ | ✅ | ✅ | ✅ | | [DINOv2](https://github.com/facebookresearch/dinov2) | Vision-Self-Supervised | [demo](examples/dinov2) | ✅ | ✅ | ✅ | ✅ |
@ -50,57 +46,50 @@ A Rust library integrated with **ONNXRuntime**, providing a collection of **Comp
| [Depth-Anything](https://github.com/LiheYoung/Depth-Anything) | Monocular Depth Estimation | [demo](examples/depth-anything) | ✅ | ✅ | ❌ | ❌ | | [Depth-Anything](https://github.com/LiheYoung/Depth-Anything) | Monocular Depth Estimation | [demo](examples/depth-anything) | ✅ | ✅ | ❌ | ❌ |
| [MODNet](https://github.com/ZHKKKe/MODNet) | Image Matting | [demo](examples/modnet) | ✅ | ✅ | ✅ | ✅ | | [MODNet](https://github.com/ZHKKKe/MODNet) | Image Matting | [demo](examples/modnet) | ✅ | ✅ | ✅ | ✅ |
## Solution Models
<details close>
<summary>Additionally, this repo also provides some solution models.</summary>
| Model | Example | Result |
| :---------------------------------------------------------------------------------------------------------: | :------------------------------: | :-----------------------------------------------------------------------------: |
| Lane Line Segmentation<br /> Drivable Area Segmentation<br />Car Detection<br />车道线-可行驶区域-车辆检测 | [demo](examples/yolov8-plastic-bag) | <img src='examples/yolop/demo.png' width="220px" height="140px"> |
| Face Parsing<br /> 人脸解析 | [demo](examples/face-parsing) | <img src='examples/face-parsing/demo.png' width="220px" height="200px"> |
| Text Detection<br />(PPOCR-det v3, v4)<br />通用文本检测 | [demo](examples/db) | <img src='examples/db/demo.png' width="250px" height="200px"> |
| Text Recognition<br />(PPOCR-rec v3, v4)<br />中英文-文本识别 | [demo](examples/svtr) | |
| Face-Landmark Detection<br />人脸 & 关键点检测 | [demo](examples/yolov8-face) | <img src='examples/yolov8-face/demo.png' width="220px" height="180px"> |
| Head Detection<br /> 人头检测 | [demo](examples/yolov8-head) | <img src='examples/yolov8-head/demo.png' width="220px" height="180px"> |
| Fall Detection<br /> 摔倒检测 | [demo](examples/yolov8-falldown) | <img src='examples/yolov8-falldown/demo.png' width="220px" height="180px"> |
| Trash Detection<br /> 垃圾检测 | [demo](examples/yolov8-plastic-bag) | <img src='examples/yolov8-trash/demo.png' width="250px" height="180px"> |
</details>
## Demo
```
cargo run -r --example yolov8 # yolov9, blip, clip, dinov2, svtr, db, yolo-world...
```
## Installation ## Installation
check **[ort guide](https://ort.pyke.io/setup/linking)** Refer to **[ort guide](https://ort.pyke.io/setup/linking)**
<details close> <details close>
<summary>For Linux or MacOS users</summary> <summary>For Linux or MacOS users</summary>
- Firstly, download from latest release from [ONNXRuntime Releases](https://github.com/microsoft/onnxruntime/releases) - Firstly, download from latest release from [ONNXRuntime Releases](https://github.com/microsoft/onnxruntime/releases)
- Then linking - Then linking
```shell ```Shell
export ORT_DYLIB_PATH=/Users/qweasd/Desktop/onnxruntime-osx-arm64-1.17.1/lib/libonnxruntime.1.17.1.dylib export ORT_DYLIB_PATH=/Users/qweasd/Desktop/onnxruntime-osx-arm64-1.17.1/lib/libonnxruntime.1.17.1.dylib
``` ```
</details> </details>
## Demo
```Shell
cargo run -r --example yolov8 # yolov10, blip, clip, yolop, svtr, db, yolo-world, ...
```
## Integrate into your own project ## Integrate into your own project
<details close> <details close>
<summary>Expand</summary> <summary>Expand</summary>
#### 1. Add `usls` as a dependency to your project's `Cargo.toml` ### 1. Add `usls` as a dependency to your project's `Cargo.toml`
```shell ```Shell
usls = { git = "https://github.com/jamjamjon/usls", rev = "xxx"} cargo add usls
``` ```
#### 2. Set `Options` and build model Or you can use specific commit
```Shell
usls = { git = "https://github.com/jamjamjon/usls", rev = "???sha???"}
```
### 2. Set `Options` and build model
```Rust ```Rust
let options = Options::default() let options = Options::default()
@ -201,5 +190,24 @@ pub struct Y {
- Other tasks results can be found at: `src/ys/y.rs` - Other tasks results can be found at: `src/ys/y.rs`
</details>
## Solution Models
<details close>
<summary>Additionally, this repo also provides some solution models.</summary>
| Model | Example | Result |
| :---------------------------------------------------------------------------------------------------------: | :------------------------------: | :-----------------------------------------------------------------------------: |
| Lane Line Segmentation<br /> Drivable Area Segmentation<br />Car Detection<br />车道线-可行驶区域-车辆检测 | [demo](examples/yolov8-plastic-bag) | <img src='examples/yolop/demo.png' width="220px" height="140px"> |
| Face Parsing<br /> 人脸解析 | [demo](examples/face-parsing) | <img src='examples/face-parsing/demo.png' width="220px" height="200px"> |
| Text Detection<br />(PPOCR-det v3, v4)<br />通用文本检测 | [demo](examples/db) | <img src='examples/db/demo.png' width="250px" height="200px"> |
| Text Recognition<br />(PPOCR-rec v3, v4)<br />中英文-文本识别 | [demo](examples/svtr) | |
| Face-Landmark Detection<br />人脸 & 关键点检测 | [demo](examples/yolov8-face) | <img src='examples/yolov8-face/demo.png' width="220px" height="180px"> |
| Head Detection<br /> 人头检测 | [demo](examples/yolov8-head) | <img src='examples/yolov8-head/demo.png' width="220px" height="180px"> |
| Fall Detection<br /> 摔倒检测 | [demo](examples/yolov8-falldown) | <img src='examples/yolov8-falldown/demo.png' width="220px" height="180px"> |
| Trash Detection<br /> 垃圾检测 | [demo](examples/yolov8-plastic-bag) | <img src='examples/yolov8-trash/demo.png' width="250px" height="180px"> |
</details> </details>

4
src/core/annotator.rs
View File

@ -109,7 +109,7 @@ impl Annotator {
self self
} }
/// Plotting BBOXes or not /// Plotting bboxes or not
pub fn without_bboxes(mut self, x: bool) -> Self { pub fn without_bboxes(mut self, x: bool) -> Self {
self.without_bboxes = x; self.without_bboxes = x;
self self
@ -195,7 +195,7 @@ impl Annotator {
self self
} }
/// Plotting MBRs or not /// Plotting mbrs or not
pub fn without_mbrs(mut self, x: bool) -> Self { pub fn without_mbrs(mut self, x: bool) -> Self {
self.without_mbrs = x; self.without_mbrs = x;
self self

15
src/core/engine.rs
View File

@ -32,7 +32,7 @@ pub struct OrtEngine {
model_proto: onnx::ModelProto, model_proto: onnx::ModelProto,
params: usize, params: usize,
wbmems: usize, wbmems: usize,
pub ts: Ts, ts: Ts,
} }
impl OrtEngine { impl OrtEngine {
@ -353,7 +353,7 @@ impl OrtEngine {
Ok(ys) Ok(ys)
} }
pub fn _set_ixx(x: isize, ixx: &Option<MinOptMax>, i: usize, ii: usize) -> Option<MinOptMax> { fn _set_ixx(x: isize, ixx: &Option<MinOptMax>, i: usize, ii: usize) -> Option<MinOptMax> {
match x { match x {
-1 => { -1 => {
match ixx { match ixx {
@ -369,7 +369,8 @@ impl OrtEngine {
} }
} }
pub fn nbytes_from_onnx_dtype_id(x: usize) -> usize { #[allow(dead_code)]
fn nbytes_from_onnx_dtype_id(x: usize) -> usize {
match x { match x {
7 | 11 | 13 => 8, // i64, f64, u64 7 | 11 | 13 => 8, // i64, f64, u64
1 | 6 | 12 => 4, // f32, i32, u32 1 | 6 | 12 => 4, // f32, i32, u32
@ -380,7 +381,8 @@ impl OrtEngine {
} }
} }
pub fn nbytes_from_onnx_dtype(x: &ort::TensorElementType) -> usize { #[allow(dead_code)]
fn nbytes_from_onnx_dtype(x: &ort::TensorElementType) -> usize {
match x { match x {
ort::TensorElementType::Float64 ort::TensorElementType::Float64
| ort::TensorElementType::Uint64 | ort::TensorElementType::Uint64
@ -399,6 +401,7 @@ impl OrtEngine {
} }
} }
#[allow(dead_code)]
fn ort_dtype_from_onnx_dtype_id(value: i32) -> Option<ort::TensorElementType> { fn ort_dtype_from_onnx_dtype_id(value: i32) -> Option<ort::TensorElementType> {
match value { match value {
0 => None, 0 => None,
@ -630,4 +633,8 @@ impl OrtEngine {
pub fn memory_weights(&self) -> usize { pub fn memory_weights(&self) -> usize {
self.wbmems self.wbmems
} }
pub fn ts(&self) -> &Ts {
&self.ts
}
} }

2
src/core/onnx.rs
View File

@ -1,3 +1,5 @@
//! ONNX file generated by prost-build.
// This file is @generated by prost-build. // This file is @generated by prost-build.
/// Attributes /// Attributes
/// ///

2
src/core/ops.rs
View File

@ -1,3 +1,5 @@
//! Some processing functions to image and ndarray.
use anyhow::Result; use anyhow::Result;
use fast_image_resize as fir; use fast_image_resize as fir;
use fast_image_resize::{ use fast_image_resize::{

2
src/core/options.rs
View File

@ -1,3 +1,5 @@
//! Options for build models.
use anyhow::Result; use anyhow::Result;
use crate::{ use crate::{

1
src/core/x.rs
View File

@ -4,6 +4,7 @@ use ndarray::{Array, Dim, IxDyn, IxDynImpl};
use crate::Ops; use crate::Ops;
/// Model input, alias for [`Array<f32, IxDyn>`]
#[derive(Debug, Clone, Default)] #[derive(Debug, Clone, Default)]
pub struct X(pub Array<f32, IxDyn>); pub struct X(pub Array<f32, IxDyn>);

107
src/lib.rs
View File

@ -1,3 +1,110 @@
//! A Rust library integrated with ONNXRuntime, providing a collection of Computer Vison and Vision-Language models.
//!
//! [`OrtEngine`] provides ONNX model loading, metadata parsing, dry_run, inference and other functions, supporting EPs such as CUDA, TensorRT, CoreML, etc. You can use it as the ONNXRuntime engine for building models.
//!
//!
//!
//!
//! # Supported models
//! | Model | Task / Type | Example | CUDA<br />f32 | CUDA<br />f16 | TensorRT<br />f32 | TensorRT<br />f16 |
//! | :---------------------------------------------------------------: | :-------------------------: | :----------------------: | :-----------: | :-----------: | :------------------------: | :-----------------------: |
//! | [YOLOv5](https://github.com/ultralytics/yolov5) | Object Detection<br />Instance Segmentation<br />Classification | [demo](examples/yolov5) | ✅ | ✅ | ✅ | ✅ |
//! | [YOLOv8-obb](https://github.com/ultralytics/ultralytics) | Object Detection<br />Instance Segmentation<br />Classification<br />Oriented Object Detection<br />Keypoint Detection | [demo](examples/yolov8) | ✅ | ✅ | ✅ | ✅ |
//! | [YOLOv9](https://github.com/WongKinYiu/yolov9) | Object Detection | [demo](examples/yolov9) | ✅ | ✅ | ✅ | ✅ |
//! | [YOLOv10](https://github.com/THU-MIG/yolov10) | Object Detection | [demo](examples/yolov10) | ✅ | ✅ | ✅ | ✅ |
//! | [RT-DETR](https://arxiv.org/abs/2304.08069) | Object Detection | [demo](examples/rtdetr) | ✅ | ✅ | ✅ | ✅ |
//! | [FastSAM](https://github.com/CASIA-IVA-Lab/FastSAM) | Instance Segmentation | [demo](examples/fastsam) | ✅ | ✅ | ✅ | ✅ |
//! | [YOLO-World](https://github.com/AILab-CVC/YOLO-World) | Object Detection | [demo](examples/yolo-world) | ✅ | ✅ | ✅ | ✅ |
//! | [DINOv2](https://github.com/facebookresearch/dinov2) | Vision-Self-Supervised | [demo](examples/dinov2) | ✅ | ✅ | ✅ | ✅ |
//! | [CLIP](https://github.com/openai/CLIP) | Vision-Language | [demo](examples/clip) | ✅ | ✅ | ✅ visual<br />❌ textual | ✅ visual<br />❌ textual |
//! | [BLIP](https://github.com/salesforce/BLIP) | Vision-Language | [demo](examples/blip) | ✅ | ✅ | ✅ visual<br />❌ textual | ✅ visual<br />❌ textual |
//! | [DB](https://arxiv.org/abs/1911.08947) | Text Detection | [demo](examples/db) | ✅ | ✅ | ✅ | ✅ |
//! | [SVTR](https://arxiv.org/abs/2205.00159) | Text Recognition | [demo](examples/svtr) | ✅ | ✅ | ✅ | ✅ |
//! | [RTMO](https://github.com/open-mmlab/mmpose/tree/main/projects/rtmo) | Keypoint Detection | [demo](examples/rtmo) | ✅ | ✅ | ❌ | ❌ |
//! | [YOLOPv2](https://arxiv.org/abs/2208.11434) | Panoptic Driving Perception | [demo](examples/yolop) | ✅ | ✅ | ✅ | ✅ |
//! | [Depth-Anything](https://github.com/LiheYoung/Depth-Anything) | Monocular Depth Estimation | [demo](examples/depth-anything) | ✅ | ✅ | ❌ | ❌ |
//! | [MODNet](https://github.com/ZHKKKe/MODNet) | Image Matting | [demo](examples/modnet) | ✅ | ✅ | ✅ | ✅ |
//! # Use provided models for inference
//! #### 1. Using provided [`models`] with [`Option`]
//! ```Rust, no_run
//! use usls::{coco, models::YOLO, Annotator, DataLoader, Options, Vision};
//!
//! let options = Options::default()
//! .with_model("yolov8m-seg-dyn.onnx")?
//! .with_trt(0)
//! .with_fp16(true)
//! .with_i00((1, 1, 4).into())
//! .with_i02((224, 640, 800).into())
//! .with_i03((224, 640, 800).into())
//! .with_confs(&[0.4, 0.15]) // class_0: 0.4, others: 0.15
//! .with_profile(false);
//! let mut model = YOLO::new(options)?;
//! ```
//! #### 2. Load images using [`DataLoader`] or [`image::io::Reader`]
//!
//! ```Rust, no_run
//! // Load one image
//! let x = vec![DataLoader::try_read("./assets/bus.jpg")?];
//!
//! // Load images with batch_size = 4
//! let dl = DataLoader::default()
//! .with_batch(4)
//! .load("./assets")?;
//! // Load one image with `image::io::Reader`
//! let x = image::io::Reader::open("myimage.png")?.decode()?
//! ```
//!
//! #### 3. Build annotator using [`Annotator`]
//!
//! ```Rust, no_run
//! let annotator = Annotator::default()
//! .with_bboxes_thickness(7)
//! .with_saveout("YOLOv8");
//! ```
//!
//! #### 4. Run and annotate
//!
//! ```Rust, no_run
//! for (xs, _paths) in dl {
//! let ys = model.run(&xs)?;
//! annotator.annotate(&xs, &ys);
//! }
//! ```
//!
//! #### 5. Parse inference results from [`Vec<Y>`]
//! For example, uou can get detection bboxes with `y.bboxes()`:
//! ```Rust, no_run
//! let ys = model.run(&xs)?;
//! for y in ys {
//! // bboxes
//! if let Some(bboxes) = y.bboxes() {
//! for bbox in bboxes {
//! println!(
//! "Bbox: {}, {}, {}, {}, {}, {}",
//! bbox.xmin(),
//! bbox.ymin(),
//! bbox.xmax(),
//! bbox.ymax(),
//! bbox.confidence(),
//! bbox.id(),
//! )
//! }
//! }
//! }
//! ```
//!
//!
//! # Build your own model with [`OrtEngine`]
//!
//! Refer to [Demo: Depth-Anything](https://github.com/jamjamjon/usls/blob/main/src/models/depth_anything.rs)
//!
//!
mod core; mod core;
pub mod models; pub mod models;
mod utils; mod utils;

2
src/models/mod.rs
View File

@ -1,3 +1,5 @@
//! Models provided: [`Blip`], [`Clip`], [`YOLO`], [`DepthAnything`], ...
mod blip; mod blip;
mod clip; mod clip;
mod db; mod db;

2
src/utils/coco.rs
View File

@ -1,3 +1,5 @@
//! Some constants releated with COCO dataset: [`SKELETONS_16`], [`KEYPOINTS_NAMES_17`], [`NAMES_80`]
pub const SKELETONS_16: [(usize, usize); 16] = [ pub const SKELETONS_16: [(usize, usize); 16] = [
(0, 1), (0, 1),
(0, 2), (0, 2),

2
src/utils/colormap256.rs
View File

@ -1,3 +1,5 @@
//! Some colormap: [`TURBO`], [`INFERNO`], [`PLASMA`], [`VIRIDIS`], [`MAGMA`], [`BENTCOOLWARM`], [`BLACKBODY`], [`EXTENDEDKINDLMANN`], [`KINDLMANN`], [`SMOOTHCOOLWARM`].
pub const TURBO: [[u8; 3]; 256] = [ pub const TURBO: [[u8; 3]; 256] = [
[48, 18, 59], [48, 18, 59],
[50, 21, 67], [50, 21, 67],

20
src/utils/mod.rs
View File

@ -8,12 +8,13 @@ pub mod colormap256;
pub use colormap256::*; pub use colormap256::*;
pub const GITHUB_ASSETS: &str = "https://github.com/jamjamjon/assets/releases/download/v0.0.1"; pub(crate) const GITHUB_ASSETS: &str =
pub const CHECK_MARK: &str = "✅"; "https://github.com/jamjamjon/assets/releases/download/v0.0.1";
pub const CROSS_MARK: &str = "❌"; pub(crate) const CHECK_MARK: &str = "✅";
pub const SAFE_CROSS_MARK: &str = "❎"; pub(crate) const CROSS_MARK: &str = "❌";
pub(crate) const SAFE_CROSS_MARK: &str = "❎";
pub fn auto_load<P: AsRef<Path>>(src: P, sub: Option<&str>) -> Result<String> { pub(crate) fn auto_load<P: AsRef<Path>>(src: P, sub: Option<&str>) -> Result<String> {
let src = src.as_ref(); let src = src.as_ref();
let p = if src.is_file() { let p = if src.is_file() {
src.into() src.into()
@ -33,6 +34,7 @@ pub fn auto_load<P: AsRef<Path>>(src: P, sub: Option<&str>) -> Result<String> {
Ok(p.to_str().unwrap().to_string()) Ok(p.to_str().unwrap().to_string())
} }
/// `download` sth from src to dst
pub fn download<P: AsRef<Path> + std::fmt::Debug>( pub fn download<P: AsRef<Path> + std::fmt::Debug>(
src: &str, src: &str,
dst: P, dst: P,
@ -77,7 +79,7 @@ pub fn download<P: AsRef<Path> + std::fmt::Debug>(
Ok(()) Ok(())
} }
pub fn string_now(delimiter: &str) -> String { pub(crate) fn string_now(delimiter: &str) -> String {
let t_now = chrono::Local::now(); let t_now = chrono::Local::now();
let fmt = format!( let fmt = format!(
"%Y{}%m{}%d{}%H{}%M{}%S{}%f", "%Y{}%m{}%d{}%H{}%M{}%S{}%f",
@ -86,7 +88,8 @@ pub fn string_now(delimiter: &str) -> String {
t_now.format(&fmt).to_string() t_now.format(&fmt).to_string()
} }
pub fn config_dir() -> PathBuf { #[allow(dead_code)]
pub(crate) fn config_dir() -> PathBuf {
match dirs::config_dir() { match dirs::config_dir() {
Some(mut d) => { Some(mut d) => {
d.push("usls"); d.push("usls");
@ -99,7 +102,8 @@ pub fn config_dir() -> PathBuf {
} }
} }
pub fn home_dir(sub: Option<&str>) -> PathBuf { #[allow(dead_code)]
pub(crate) fn home_dir(sub: Option<&str>) -> PathBuf {
match dirs::home_dir() { match dirs::home_dir() {
Some(mut d) => { Some(mut d) => {
d.push(".usls"); d.push(".usls");

2
src/ys/bbox.rs
View File

@ -1,4 +1,4 @@
/// Bounding Box 2D /// Bounding Box 2D.
#[derive(Clone, PartialEq, PartialOrd)] #[derive(Clone, PartialEq, PartialOrd)]
pub struct Bbox { pub struct Bbox {
x: f32, x: f32,

2
src/ys/embedding.rs
View File

@ -3,7 +3,7 @@ use ndarray::{Array, Axis, Ix2, IxDyn};
use crate::X; use crate::X;
/// Embedding /// Embedding for image or text.
#[derive(Clone, PartialEq, Default)] #[derive(Clone, PartialEq, Default)]
pub struct Embedding(Array<f32, IxDyn>); pub struct Embedding(Array<f32, IxDyn>);

2
src/ys/keypoint.rs
View File

@ -1,6 +1,6 @@
use std::ops::{Add, Div, Mul, Sub}; use std::ops::{Add, Div, Mul, Sub};
/// Keypoint 2D /// Keypoint 2D.
#[derive(PartialEq, Clone)] #[derive(PartialEq, Clone)]
pub struct Keypoint { pub struct Keypoint {
x: f32, x: f32,

1
src/ys/mask.rs
View File

@ -1,5 +1,6 @@
use image::DynamicImage; use image::DynamicImage;
/// Gray-Scale Mask.
#[derive(Clone, PartialEq)] #[derive(Clone, PartialEq)]
pub struct Mask { pub struct Mask {
mask: DynamicImage, mask: DynamicImage,

2
src/ys/mbr.rs
View File

@ -1,6 +1,6 @@
use geo::{coord, line_string, Area, BooleanOps, Coord, EuclideanDistance, LineString, Polygon}; use geo::{coord, line_string, Area, BooleanOps, Coord, EuclideanDistance, LineString, Polygon};
/// Minimum Bounding Rectangle /// Minimum Bounding Rectangle.
#[derive(Clone, PartialEq)] #[derive(Clone, PartialEq)]
pub struct Mbr { pub struct Mbr {
ls: LineString, ls: LineString,

1
src/ys/polygon.rs
View File

@ -5,6 +5,7 @@ use geo::{
use crate::{Bbox, Mbr}; use crate::{Bbox, Mbr};
/// Polygon.
#[derive(Clone, PartialEq)] #[derive(Clone, PartialEq)]
pub struct Polygon { pub struct Polygon {
polygon: geo::Polygon, polygon: geo::Polygon,

2
src/ys/prob.rs
View File

@ -1,4 +1,4 @@
/// Probabilities for classification /// Probabilities for classification.
#[derive(Clone, PartialEq, Default)] #[derive(Clone, PartialEq, Default)]
pub struct Prob { pub struct Prob {
probs: Vec<f32>, probs: Vec<f32>,

1
src/ys/y.rs
View File

@ -1,5 +1,6 @@
use crate::{Bbox, Embedding, Keypoint, Mask, Mbr, Polygon, Prob}; use crate::{Bbox, Embedding, Keypoint, Mask, Mbr, Polygon, Prob};
/// Inference results container for each image.
#[derive(Clone, PartialEq, Default)] #[derive(Clone, PartialEq, Default)]
pub struct Y { pub struct Y {
probs: Option<Prob>, probs: Option<Prob>,