I will build custom yolo object detection and segmentation models in python
About this Gig
Need a computer vision model that works on your own images? I will train and deliver a custom YOLO model for object detection or segmentation, ready to use in your project.
What I can build:
- Object detection and counting (products, vehicles, people, defects)
- Image segmentation with YOLO-seg or SAM2
- OCR for text, labels and number plates
- Detection and tracking on video or webcam
What you get:
- A trained model (PyTorch or ONNX) tuned to your data
- Accuracy report (mAP, precision, recall) with sample results
- Clean Python code to run it on images, video or webcam
- FastAPI endpoint or simple web demo (Premium)
- Free fixes for 7 days after delivery
No labeled data yet? I can label a small dataset for you or start from a pretrained model.
Tools: Python, Ultralytics YOLO, SAM2, OpenCV, PyTorch, FastAPI.
Send me a few sample images and your goal, and I will recommend the right package before you order.
Expertise:
Image processing
•
Classification
•
Object detection
Programming language:
Python
Tools:
Jupyter Notebook
•
OpenCV
•
PyTorch
Frameworks:
PyTorch
FAQ
Do I need labeled data?
It helps, but it is not required. I can label a small dataset for you or start from a pretrained model. Large labeling jobs are quoted separately.
Can it run without a GPU?
Yes. I can export a lightweight ONNX model that runs on a normal computer. I handle the training on my side.
Can you detect my own custom objects?
Yes, that is what the Standard package is for. Send me sample images and I will confirm before you order.
Do I get the model and source code?
Yes. You receive the trained model weights and the full Python code.

