I will build a custom object detection model with yolo and computer vision
AI and ML Engineer, AI Chatbots, AI Agents and Computer Vision
About this Gig
Need a model that can find specific objects in your images or video? I train custom YOLO object detection models on your own data and hand over something you can actually run.
What I can build:
- Detection for products, vehicles, people, defects, safety gear or anything you have examples of
- - Image and video processing, including camera feeds
- - A FastAPI endpoint so your app can call the model
- - Docker setup so it is easy to deploy
Recent work: my final year project detects road accidents and fire from video in real time. The YOLOv12 model reached 89.9% mAP@50 and 92.5% precision on 1,800 test images.
How it works:
- You share your images or video and the objects you need
- 2. I check the data and tell you honestly if it is enough
- 3. I clean or label the data, train, test and send you the results
- 4. You get the trained model, the code and clear instructions
Please message me before ordering so we can agree on the classes and data size.
Programming language:
Python
•
Colab
Tools:
Jupyter Notebook
•
OpenCV
•
CVAT
•
PyTorch
Frameworks:
Scikit-learn
•
PyTorch
•
Panda
FAQ
How many images do I need?
Usually a few hundred labeled examples per object is a good start. If you have less, send what you have and I will tell you honestly whether it is enough or how to collect more.
Can it work on live video or a webcam?
Yes. I can set the model up to run on video files, a webcam or a camera stream. Speed depends on your hardware, so I will test it and share the frames per second you can expect.

