Yolo-FastestV2 - GitLite

KO | EN

gitlite — search

#javascript #ai #llm #go #machine-learning #rust #docker #nodejs #awesome #cli #pytorch #mcp #ios #openai

Yolo-FastestV2

(★ 936)

:zap: Based on Yolo's low-power, ultra-lightweight universal target detection algorithm, the parameter is only 250k, and the speed of the smart phone mobile terminal can reach ~300fps+

파일 탐색기

최종 버전 다운로드 (.zip)

- coco.data
- coco.names
- 000004.jpg
- 000004_result.png
- 000139.jpg
- 000139_result.png
- 000148.jpg
- 000148_result.png
- 000181.jpg
- 000181_result.png
- 000230.jpg
- 000230_result.png
- demo.png
- - backbone.pth
  - shufflenetv2.py
- detector.py
- fpn.py
- coco2017-0.241078ap-model.pth
- - build.sh
  - demo.cpp
  - output.png
  - test.jpg
- datasets.py
- loss.py
- utils.py
- README.md
.gitignore
evaluation.py
genanchors.py
pytorch2onnx.py
README.md
requirements.txt
test.py
train.py

# 설치 가이드

1. 코드 내려받기

git clone https://github.com/dog-qiuqiu/Yolo-FastestV2

깃허브에서 프로젝트 코드 전체를 내 컴퓨터로 내려받습니다.

cd Yolo-FastestV2

방금 내려받은 프로젝트 폴더 안으로 이동합니다.

2. Python

쉬움 추천

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
Python 3 설치 시 'Add Python to PATH' 옵션을 꼭 체크하세요 (Windows).

pip3 install -r requirements.txt

requirements.txt 등에 명시된 파이썬 라이브러리를 설치합니다.

python3 -m onnxsim yolo-fastestv2.onnx yolo-fastestv2-opt.onnx

파이썬 스크립트(또는 모듈)를 실행합니다.

✅ 에러 메시지 없이 실행되고 터미널에 안내 문구가 출력되면 정상입니다.

이 레포의 README에 적힌 실제 명령어를 그대로 가져왔습니다.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# ***FastestDet: it has higher accuracy and faster speed than Yolo-fastest https://github.com/dog-qiuqiu/FastestDet***
# :zap:Yolo-FastestV2:zap:[![DOI](https://zenodo.org/badge/386585431.svg)](https://zenodo.org/badge/latestdoi/386585431)
![image](https://github.com/dog-qiuqiu/Yolo-FastestV2/blob/main/img/demo.png)
* ***Simple, fast, compact, easy to transplant***
* ***Less resource occupation, excellent single-core performance, lower power consumption***
* ***Faster and smaller:Trade 0.3% loss of accuracy for 30% increase in inference speed, reducing the amount of parameters by 25%***
* ***Fast training speed, low computing power requirements, training only requires 3GB video memory, gtx1660ti training COCO 1 epoch only takes 4 minutes***
* ***算法介绍：https://zhuanlan.zhihu.com/p/400474142 交流qq群:1062122604***
# Evaluating indicator/Benchmark
Network|COCO mAP(0.5)|Resolution|Run Time(4xCore)|Run Time(1xCore)|FLOPs(G)|Params(M)
:---:|:---:|:---:|:---:|:---:|:---:|:---:
[Yolo-FastestV2](https://github.com/dog-qiuqiu/Yolo-FastestV2/tree/main/modelzoo)|24.10 %|352X352|3.29 ms|5.37 ms|0.212|0.25M
[Yolo-FastestV1.1](https://github.com/dog-qiuqiu/Yolo-Fastest/tree/master/ModelZoo/yolo-fastest-1.1_coco)|24.40 %|320X320|4.23 ms|7.54 ms|0.252|0.35M
[Yolov4-Tiny](https://raw.githubusercontent.com/AlexeyAB/darknet/master/cfg/yolov4-tiny.cfg)|40.2%|416X416|26.00ms|55.44ms|6.9|5.77M

* ***Test platform Mate 30 Kirin 990 CPU，Based on [NCNN](https://github.com/Tencent/ncnn)***
# Improvement
* Different loss weights for different scale output layers
* The backbone is replaced with a more lightweight shufflenetV2
* Anchor matching mechanism and loss are replaced by YoloV5, and the classification loss is replaced by softmax cross entropy from sigmoid
* Decouple the detection head, distinguish obj (foreground background classification), cls (category classification), reg (detection frame regression) 3 branches,  
# How to use
## Dependent installation
  * PIP
  ```
  pip3 install -r requirements.txt
  ```
## Test
* Picture test
  ```
  python3 test.py --data data/coco.data --weights modelzoo/coco2017-0.241078ap-model.pth --img img/000139.jpg
  ```
<div align=center>
<img src="https://github.com/dog-qiuqiu/Yolo-FastestV2/blob/main/img/000139_result.png"> />
</div>

## How to train
### Building data sets(The dataset is constructed in the same way as darknet yolo)
* The format of the data set is the same as that of Darknet Yolo, Each image corresponds to a .txt label file. The label format is also based on Darknet Yolo's data set label format: "category cx cy wh", where category is the category subscript, cx, cy are the coordinates of the center point of the normalized label box, and w, h are the normalized label box The width and height, .txt label file content example as follows:
  ```
  11 0.344192634561 0.611 0.416430594901 0.262
  14 0.509915014164 0.51 0.974504249292 0.972
  ```
* The image and its corresponding label file have the same name and are stored in the same directory. The data file structure is as follows:
  ```
  .
  ├── train
  │   ├── 000001.jpg
  │   ├── 000001.txt
  │   ├── 000002.jpg
  │   ├── 000002.txt
  │   ├── 000003.jpg
  │   └── 000003.txt
  └── val
      ├── 000043.jpg
      ├── 000043.txt
      ├── 000057.jpg
      ├── 000057.txt
      ├── 000070.jpg
      └── 000070.txt
  ```
* Generate a dataset path .txt file, the example content is as follows：
  
  train.txt
  ```
  /home/qiuqiu/Desktop/dataset/train/000001.jpg
  /home/qiuqiu/Desktop/dataset/train/000002.jpg
  /home/qiuqiu/Desktop/dataset/train/000003.jpg
  ```
  val.txt
  ```
  /home/qiuqiu/Desktop/dataset/val/000070.jpg
  /home/qiuqiu/Desktop/dataset/val/000043.jpg
  /home/qiuqiu/Desktop/dataset/val/000057.jpg
  ```
* Generate the .names category label file, the sample content is as follows:
 
  category.names
  ```
  person
  bicycle
  car
  motorbike
  ...
  
  ```
* The directory structure of the finally constructed training data set is as follows:
  ```
  .
  ├── category.names        # .names category label file
  ├── train                 # train dataset
  │   ├── 000001.jpg
  │   ├── 000001.txt
  │   ├── 000002.jpg
  │   ├── 000002.txt
  │   ├── 000003.jpg
  │   └── 000003.txt
  ├── train.txt              # train dataset path .txt file
  ├── val                    # val dataset
  │   ├── 000043.jpg
  │   ├── 000043.txt
  │   ├── 000057.jpg
  │   ├── 000057.txt
  │   ├── 000070.jpg
  │   └── 000070.txt
  └── val.txt                # val dataset path .txt file

```
### Get anchor bias
* Generate anchor based on current dataset
  ```
  python3 genanchors.py --traintxt ./train.txt
  ```
* The anchors6.txt file will be generated in the current directory,the sample content of the anchors6.txt is as follows:
  ```
  12.64,19.39, 37.88,51.48, 55.71,138.31, 126.91,78.23, 131.57,214.55, 279.92,258.87  # anchor bias
  0.636158                                                                             # iou
  ```
### Build the training .data configuration file
* Reference./data/coco.data
  ```
  [name]
  model_name=coco           # model name

[train-configure]
  epochs=300                # train epichs
  steps=150,250             # Declining learning rate steps
  batch_size=64             # batch size
  subdivisions=1            # Same as the subdivisions of the darknet cfg file
  learning_rate=0.001       # learning rate

[model-configure]
  pre_weights=None          # The path to load the model, if it is none, then restart the training
  classes=80                # Number of detection categories
  width=352                 # The width of the model input image
  height=352                # The height of the model input image
  anchor_num=3              # anchor num
  anchors=12.64,19.39, 37.88,51.48, 55.71,138.31, 126.91,78.23, 131.57,214.55, 279.92,258.87 #anchor bias

[data-configure]
  train=/media/qiuqiu/D/coco/train2017.txt   # train dataset path .txt file
  val=/media/qiuqiu/D/coco/val2017.txt       # val dataset path .txt file 
  names=./data/coco.names                    # .names category label file
  ```
### Train
* Perform training tasks
  ```
  python3 train.py --data data/coco.data
  ```
### Evaluation
* Calculate map evaluation
  ```
  python3 evaluation.py --data data/coco.data --weights modelzoo/coco2017-0.241078ap-model.pth
  ```
# Deploy
## NCNN
* Convert onnx
  ```
  python3 pytorch2onnx.py --data data/coco.data --weights modelzoo/coco2017-0.241078ap-model.pth --output yolo-fastestv2.onnx
  ```
* onnx-sim
  ```
  python3 -m onnxsim yolo-fastestv2.onnx yolo-fastestv2-opt.onnx
  ```
* Build NCNN
  ```
  git clone https://github.com/Tencent/ncnn.git
  cd ncnn
  mkdir build
  cd build
  cmake ..
  make
  make install
  cp -rf ./ncnn/build/install/* ~/Yolo-FastestV2/sample/ncnn
  ```
* Covert ncnn param and bin
  ```
  cd ncnn/build/tools/onnx
  ./onnx2ncnn yolo-fastestv2-opt.onnx yolo-fastestv2.param yolo-fastestv2.bin
  cp yolo-fastestv2* ../
  cd ../
  ./ncnnoptimize yolo-fastestv2.param yolo-fastestv2.bin yolo-fastestv2-opt.param yolo-fastestv2-opt.bin 1
  cp yolo-fastestv2-opt* ~/Yolo-FastestV2/sample/ncnn/model
  ```
* run sample
  ```
  cd ~/Yolo-FastestV2/sample/ncnn
  sh build.sh
  ./demo
  ```
# Reference
* https://github.com/Tencent/ncnn
* https://github.com/AlexeyAB/darknet
* https://github.com/ultralytics/yolov5
* https://github.com/eriklindernoren/PyTorch-YOLOv3