Pose recognition game controls
PoseControl leverages MediaPipe Pose and OpenCV to create game controls based on real-time pose recognition. This project detects specific body movements to trigger game actions such as attacking, blocking, and dodging.
- Real-time Pose Detection: Uses MediaPipe Pose for accurate and efficient pose estimation.
- Game Controls Integration: Maps specific body movements to game actions like attack, block, and dodge.
- Multi-Action Recognition: Detects arm hits, blocks, and dodges based on pose angles and visibility.
- Webcam Input: Captures input from a webcam to track and process user movements.
Clone the repository and install the required dependencies:
git clone https://github.com/yourusername/PoseControl.git
cd PoseControl
pip install -r requirements.txtTo run the pose recognition and game control script, execute the following command:
bash
python posecontrol.py
Make sure your webcam is connected and properly configured.
The project uses the following libraries:
cv2 (OpenCV) for capturing video input and processing images.
mediapipe for pose detection and landmark recognition.
numpy for numerical calculations.
pydirectinput for simulating keyboard inputs.
math for mathematical calculations.