Browse categories
Explore
Fiverr Pro
English
$
USD
I will build a yolo object detection and tracking system in python
About this Gig
Need to detect vehicles, track objects, or count people in recorded video?
I'm Shazim. I build YOLO object detection and tracking systems using Python, OpenCV and PyTorch, with runnable code and clear setup instructions.
Basic: Detection with boxes and labels on one video up to 5 minutes.
Standard: Detection, tracking IDs and CSV results on one video up to 10 minutes.
Premium: Standard features plus one counting line and a local API to process video and return results.
Every package includes:
- Commented Python source code
- An annotated sample video
- Dependency list and setup guide for one agreed operating system
- Testing and configuration on your footage
Packages use one pretrained model and up to 3 supported object classes. Custom training, annotation, dashboards and cloud deployment are quoted separately.
Accuracy and speed depend on your footage and hardware.
Message me with a short video sample, the objects to detect and your expected output. I'll check suitability and help you choose a package.
Programming language:
Python
Tools:
Jupyter Notebook
•
OpenCV
•
TensorFlow
•
Colab
•
PyTorch
Frameworks:
Scikit-learn
•
DeepPy
•
Keras
•
PyTorch
•
Panda
My Portfolio
Other Data Science & ML Services I Offer
FAQ
What should I send before ordering?
Send a short sample video, the objects you want to detect, your required outputs, and your computer's operating system and CPU/GPU details. Include the video's resolution and duration. I'll review suitability and confirm the scope before you order
Which objects can you detect?
Packages cover up to 3 object classes already supported by one pretrained YOLO model, such as people, cars or trucks. Send your class list and footage first. Objects outside the model's supported classes may require separately quoted custom training
Is custom model training included?
The listed packages use pretrained YOLO models. Custom training is quoted separately after reviewing your labeled dataset, target classes and available computing resources. Dataset annotation and paid computing costs require a separate agreement.
What files will I receive?
All packages include commented Python code, a dependency list, setup instructions and an annotated sample video. Standard adds tracking IDs and CSV results. Premium adds one counting line and a local API. Setup covers one operating system agreed before ordering.
Can I use the code on more videos after delivery?
Yes. The 5-minute Basic and 10-minute Standard/Premium limits apply to the sample video I process and validate. You receive reusable code for compatible videos. New camera conditions or object classes may need further testing or separately scoped changes.
What accuracy and processing speed can I expect?
Results depend on lighting, camera angle, object size, occlusion, video resolution and hardware. I'll assess your sample and discuss realistic expectations. Specific accuracy or FPS targets need testing on representative footage and the intended computer.
What does the Premium API include?
One local API endpoint that accepts a recorded video and returns processing results, with setup instructions and an example request. Website integration, a dashboard, public hosting, authentication and ongoing server maintenance require separate scope.
Are live cameras and cloud deployment included?
The packages cover recorded video processed on one agreed computer. For webcams, RTSP/CCTV streams, multiple cameras or cloud deployment, message me first so I can assess compatibility and quote the additional work. Hosting fees are separate.
What counts as a revision?
A revision is one round of adjustments within the agreed scope, such as output formatting, thresholds or counting-line placement. Basic includes 1; Standard and Premium include 2. An extra round costs $20 and adds 2 days. Fixes to meet agreed requirements are included.
