| Network | mAPval 0.5 | mAPval 0.5:0.95 | Resolution | Run Time(4xCore) | Run Time(1xCore) | Params(M) |
|---|---|---|---|---|---|---|
| yolov5s | 56.8% | 37.4% | 640X640 | 395.31ms | 1139.16ms | 7.2M |
| yolov6n | - | 30.8% | 416X416 | 109.24ms | 445.44ms | 4.3M |
| yolox-nano | - | 25.8% | 416X416 | 76.31ms | 191.16ms | 0.91M |
| nanodet_m | - | 20.6% | 320X320 | 49.24ms | 160.35ms | 0.95M |
| yolo-fastestv1.1 | 24.40% | - | 320X320 | 26.60ms | 75.74ms | 0.35M |
| yolo-fastestv2 | 24.10% | - | 352X352 | 23.8ms | 68.9ms | 0.25M |
| FastestDet | 25.3% | 13.0% | 352X352 | 23.51ms | 70.62ms | 0.24M |
- Test platform Radxa Rock3A RK3568 ARM Cortex-A55 CPU，Based on NCNN
- CPU lock frequency 2.0GHz
- Anchor-Free
- Single scale detector head
- Cross grid multiple candidate targets
- Dynamic positive and negative sample allocation
| Equipment | Computing backend | System | Framework | Run time(Single core) | Run time(Multi core) |
|---|---|---|---|---|---|
| Radxa rock3a | RK3568(arm-cpu) | Linux(aarch64) | ncnn | 70.62ms | 23.51ms |
| Radxa rock3a | RK3568(NPU) | Linux(aarch64) | rknn | 28ms | - |
| Qualcomm | Snapdragon 835(arm-cpu) | Android(aarch64) | ncnn | 32.34ms | 16.24ms |
| Intel | i7-8700(X86-cpu) | Linux(amd64) | ncnn | 4.51ms | 4.33ms |
- PiP(Note pytorch CUDA version selection)
pip install -r requirements.txt
- Picture test
python3 test.py --yaml configs/coco.yaml --weight weights/weight_AP05:0.253207_280-epoch.pth --img data/3.jpg
/>
-
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
./train/000001.jpg ./train/000002.jpg ./train/000003.jpgval.txt
./val/000070.jpg ./val/000043.jpg ./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