I will build ai cctv analytics for ppe, fall detection and people counting

Vietnam

I speak English, Vietnamese

AI Engineer for Computer Vision and LLM Apps

I build AI systems that work in the real world. My recent projects include SYL GuardianCam, a real-time fall detection system (YOLO Pose, FastAPI, Docker) with 83% F1 that placed Top 10 at AI20K, and ...
About this Gig

Want your existing cameras to flag missing helmets, detect falls or count people? I build computer vision pipelines that run on real CCTV and RTSP streams, not only in a notebook.


What I have built:

  • SYL GuardianCam: real-time fall and immobility detection from RTSP cameras (YOLO Pose, multi-person tracking, ONNX Runtime, FastAPI WebSocket, React dashboard, Docker Compose). 83.05% F1 on a 100-video benchmark. Top 10 Product Showcase at AI20K (Vingroup x VinUni).
  • PPE detection (helmet, vest, gloves, eye protection): merged 5 public datasets into 6,956 clean images and compared 10+ methods. mAP50 0.674 to 0.753 on an 808-image test set.
  • Defect-classification models on factory data (Nidec internship).

What I can build for you:

  • PPE compliance alerts per zone
  • Fall / person-down detection
  • People or vehicle counting (tracking + line crossing)
  • Zone intrusion and dwell-time alerts
  • ONNX export for CPU or edge GPU, FastAPI API, Docker deployment

Start small: the $90 demo runs a model on your own 1-minute clip, so you see real accuracy before you commit.


Please message me first with a sample frame and what you need detected.

Expertise:

Image processing

•

Classification

•

Object detection

Programming language:

Python

•

SQL

Tools:

OpenCV

•

PyTorch

Frameworks:

PyTorch

•

Other