Browse categories
Explore
Fiverr Pro
English
$
USD
I will train a custom yolo object detection model on your data
Computer Vision Engineer: Video Analytics For Surveillance, Safety and Sports
I Deliver Production-Ready Computer Vision systems for Real-time video analytics, Object Detection, Pose estimation, Action Recognition, and Advanced AI pipelines. I build fast, accurate models using ...
Level 1
Has met certain performance criteria and shows strong potential in the marketplace.
About this Gig
Need a custom AI model to detect, count or track objects in your images or video? I train production-ready YOLO models on YOUR data real working models, not generic demos.
What I can detect:
- Surveillance & safety people, PPE, intrusion
- Retail & inventory products, theft, shelves
- Traffic vehicles, plates, counting
- Sports player & ball tracking
- Manufacturing defects, parts
- Any custom object you need
What you get:
- Custom YOLO model trained on your dataset
- Accuracy & evaluation report
- Clean source code + inference script
- Optional cloud/edge deployment + API
- Documentation so you can run it yourself
Why me: I'm a Computer Vision engineer who ships real-time detection systems running on live cameras across surveillance, retail and sports. You get a working model, fast, with honest accuracy numbers and clean code.
Stack: YOLOv8/v11, PyTorch, OpenCV, TensorFlow, Docker, GCP & AWS.
New to this? Message me with a few sample images and what you want to detect I'll tell you exactly what's possible before you order.
Programming language:
Python
•
SQL
•
Colab
•
MLflow
•
Amazon SageMaker
Frameworks:
Scikit-learn
•
Google ML Kit
•
Keras
•
PyTorch
•
Panda
My Portfolio
FAQ
I don't have a labeled dataset, can you still help?
Yes. Send your raw images or video and I'll handle the annotation as part of the project. We can also start with the POC package to prove accuracy on a small set first.
What accuracy can I expect?
It depends on your data and object type, but I optimize for the best real-world accuracy and always share honest metrics (mAP, precision, recall) so you know exactly what you're getting.
Will it run in real time on my camera or device?
Yes, I build models for live video and can deploy to cloud or edge devices (Jetson, GPU servers) with an API. Just tell me your target hardware.
Can I run the model myself after delivery?
Absolutely. You get clean, documented source code plus an inference script and setup help, so you can run it on your own.
Is my data kept confidential?
100%. Your data and project stay private, and I'm happy to sign an NDA on request.
