Kateryna Romanchuk (SDET)

Hey everyone! I've recently dived into the fascinating world of MediaPipe, focusing on two intriguing aspects: Pose Landmark Detection and Hand Landmarks Detection. The potential applications are vast, and I'm thrilled to share my initial ideas: 1️⃣ Posture Alert System : I'm developing a script that will alert me if I've been sitting for too long. This could be a game-changer for maintaining better posture and health during those long work sessions. https://github.com/safo-bora/ML-pet/blob/main/mediapipe_debug_pose.py 2️⃣ Hand Detection for Virtual Music : Just for fun (for now), I'm working on a script that turns hand movements into piano playing. Each section of the screen corresponds to a different musical tone, allowing me to play music without a physical piano. https://github.com/safo-bora/ML-pet/blob/main/mediapipe_debug_piano.py Stay tuned for more updates and feel free to share your thoughts or similar projects you're working on! 💖 Backstage:

Hi! Today, I want to delve into something truly exciting in the world of computer vision: MediaPipe. Developed by Google, MediaPipe is a framework that has been revolutionizing the way we interact with and understand visual data.  There are 11 different modules -  https://developers.google.com/mediapipe drawing_styles: Customize styles for drawing detected objects and landmarks. drawing_utils: Utility functions for drawing on images and videos. face_detection: Detect and locate faces within images and video streams. face_mesh: Analyze the contours and features of a face. face_mesh_connections: Define the connections between face landmarks. hands: Track and analyze hand gestures and positions. hands_connections: Connect landmarks in the hands for visualization. holistic: Comprehensive solution for analyzing face, hands, and body posture. objectron: 3D object detection and spatial understanding. pose: Detect and analyze human body postures. selfie_segmentation: Segment a ...