Browse categories
Explore
Fiverr Pro
English
$
USD
I will build machine learning, deep learning, computer vision, ai apps
About this Gig
I will build a custom AI model that detects the objects you care about in images, video or on a live camera, and marks each one in real time.
What I do
- Prepare and label your images
- Train a custom detection model (YOLO) on your objects
- Test it and improve accuracy on the cases it gets wrong
- Deliver the trained model, source code and instructions to run it
What it can detect
Products on a shelf, people, vehicles, safety equipment, tools, defects, and almost any object you can photograph.
Why me
I am a software developer with 14+ years of experience, so you get a working system and not only a model file.
APIs:
Microsoft Computer Vision AI
•
Google Cloud Vision API
Programming language:
Python
•
SQL
•
Colab
Tools:
Jupyter Notebook
•
OpenCV
•
TensorFlow
•
CVAT
•
Colab
•
PyTorch
Frameworks:
Scikit-learn
•
Keras
•
PyTorch
•
Panda
My Portfolio
FAQ
What do you need from me to start?
A short description of what the model should detect and where it will be used, plus your images or video if you have them. If you have no data, choose Premium and I'll collect it.
How many images do I need?
At least 150 per object type for a working model, and more for higher accuracy. Images should show the object in the conditions where it will be used: the same lighting, angles and backgrounds.
How accurate will the model be?
It depends on the data. I test the model on images it hasn't seen and report the results, then improve the cases it gets wrong. No model is 100% accurate, and I'll tell you honestly what to expect before you order.
Can it run on my camera, phone or device?
Yes in most cases. It runs on a computer with a webcam or IP camera, and I can export it for mobile or small devices. Tell me your setup before ordering so I can confirm.
Do I get the source code and the model?
Yes. Every package includes the trained model file and the source code, and they are yours to use.
