I will build real time pose estimation and action recognition system using mediapipe

V
visionbyrehman
V
visionbyrehman
Rehman A.

About this gig

I will build a real-time pose estimation and action recognition system using Computer Vision, MediaPipe, OpenCV, and Deep Learning. The system can detect human body movements, track poses, recognize actions/exercises, and perform repetition counting from webcam or video input. Suitable for fitness AI, motion analysis, sports analytics, surveillance, and smart vision applications.

Get to know Rehman A.

Rehman A.

Computer Vision Engineer

  • FromPakistan
  • Member sinceOct 2025
  • Languages

    English
Computer Vision & AI Engineer with 2+ years of experience building real-time, production-ready vision systems. I develop end-to-end solutions including data pipelines, model training, optimization, and deployment. Skilled in object detection, tracking, classification, face recognition, action recognition, and video analytics for surveillance and automation. I also build Vision-Language Models (VLMs) and multimodal AI systems (CV + NLP). I deliver scalable solutions with APIs, dashboards, and cloud deployment using AWS & Docker, fully integrated into your workflow.

My Portfolio