基于YOLOv8的摔倒行为检测系统（Python源码+Pyqt6界面）

# YoloSide ~ A GUI For YOLOv8 `V2.0-Beta` ---  ## How to Run - `python>=3.8` - `pip install ultralytics` - `pip install pyside6` - `pip install chardet` - `conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia` - `python main.py` #### !!The corresponding version of the library must be installed ## Notice - `ultralytics` follows the `GPL-3.0`, if you need commercial use, you need to obtain its license. - The tracking function is not yet complete. - If you expect to use your own model, you need to use `ultralytics` to train the yolov8/5 model first, and then put the trained `.pt` file into the `models/*` folder. - There are still some bugs in the software, and I will continue to optimize and add some more interesting functions as my time allows. - If you check the save results, they will be saved in the `./run` path - The UI design file is `home.ui`, if you modify it, you need to use the `pyside6-uic home.ui > ui/home.py` command to regenerate the `.py` file - The resource file is `resources.qrc`, if you modify the default icon, you need to use the `pyside6-rcc resoures.qrc > ui/resources_rc.py` command to regenerate the `.py` file ## To Do - [ ] The input source supports camera(if you need this function urgently, you can modify it according to the `chose_rtsp`、`load_rtsp` function in `mian.py`) - [ ] Graph showing changes in target quantity - [ ] Target tracking - [ ] Instance segmentation - [ ] Monitor system hardware usage ## References - [PyQt5-YOLOv5](https://github.com/Javacr/PyQt5-YOLOv5) - [ultralytics](https://github.com/ultralytics/ultralytics) - [PySide6-YOLOv8](https://github.com/Jai-wei/YOLOv8-PySide6-GUI/tree/main)