Browse categories
Explore
Fiverr Pro
English
$
USD
I will build human pose estimation, sports, body keypoint detection system on mediapipe
DATA SCIENCE, COMPUTER VISION, MACHINE LEARNING, PYTHON
Professional Computer Vision and Deep Learning engineer with expertise in YOLOv8, OpenCV, Python AI, MediaPipe, and image processing systems. I develop real-time object detection, tracking, face recog...
About this Gig
Track human movement with mathematical precision. I will build an advanced human pose estimation and body keypoint detection system using robust frameworks like MediaPipe or OpenPose. This system translates raw visual human movement into structured coordinate analytics.
I can map, extract, and monitor complete skeletal frameworks, tracking critical nodes across the hands, face, and body. I write clean tracking logic to extract raw (x, y, z) coordinates from video files, real-time cameras, or network streams.
Deliverables include coordinate extraction pipelines, mathematical calculation scripts to evaluate real-time joint angles, and low-latency rendering overlays. This setup is perfect for tracking body alignment or extracting geometric skeletal data. Message me today to develop your custom pose tracking engine!
Programming language:
Python
•
R
•
MATLAB
•
SQL
•
Colab
Frameworks:
Scikit-learn
•
DeepPy
•
Google ML Kit
•
Keras
•
PyTorch
Other Data Science & ML Services I Offer
FAQ
What is the benefit of MediaPipe over OpenPose?
MediaPipe is exceptionally lightweight and ideal for real-time mobile/CPU platforms, whereas OpenPose offers excellent multi-person precision at the expense of requiring powerful GPUs.
Can this system track individual finger joints?
Yes, I can initialize full holistic tracking models that capture facial landmarks, body poses, and individual hand/finger keypoints.
What format is the extracted coordinate data saved in?
The coordinate data streams can be exported directly to CSV, JSON, or broadcast live over WebSockets.
