Browse categories
Explore
Fiverr Pro
English
$
USD
I will build ai computer vision, object detection, tracking and player tracking
Multitasker
I'm Dheeraj Kumar, a full-stack AI/ML Engineer building real applied-AI systems end to end: data or video to model or LLM to API to dashboard, deployed and production-ready. Core focus: computer visio...
About this Gig
Got footage, images, or camera feeds sitting there with no system actually reading what's in them?
I am Full-stack AI/ML engineer Expert in developing computer vision systems based on AI, machine learning, object detection, tracking, OCR, image recognition & sports analytics like player & ball tracking, based on OpenCV, YOLO & Deep Learning.
WHAT YOU'LL GET
Object detection & tracking system
Multi-camera video analytics setup
Player tracking & ball detection
OCR & text extraction from images
Image recognition & classification model
Image segmentation solution
YOLO-based detection pipeline
OpenCV Python processing scripts
Motion & anomaly detection
Heatmap & movement pattern generation
WORKFLOW
Requirements Review -> Data/Footage Review -> Model Selection -> Training/Fine-tuning -> Detection & Tracking Build -> Testing & Accuracy Check -> Revisions -> Delivery
WHY CHOOSE ME
REAL PROJECTS:
1. 50+ camera CCTV tracking with a live dashboard,
2. 20,000+ image classification
3. 92% accuracy face emotion detection.
On-time delivery
100% client satisfaction guaranteed
CONTACT ME NOW and send me your data or footage
APIs:
Other
Programming language:
Python
Tools:
Jupyter Notebook
•
OpenCV
•
TensorFlow
•
CVAT
•
Colab
•
PyTorch
Frameworks:
Scikit-learn
•
Keras
•
PyTorch
My Portfolio
Other Data Science & ML Services I Offer
FAQ
Will the model actually be deployed, or left as a Jupyter notebook?
A working FastAPI backend for model deployment is included, so you receive a usable pipeline, not just code sitting in a notebook file
Can the model be trained on my own data specifically?
Yes, model fine-tuning on your own data is included, so detection and tracking are calibrated to your actual footage or images.
Is this proven at scale, or just tested on small demo datasets?
Real production builds back this, including 50+ camera CCTV tracking with a live dashboard and 20,000+ image classification projects.
Can you build player and ball tracking for sports specifically?
Yes, player tracking and ball detection for sports analytics is a specialty, alongside general object detection and tracking systems.
Do you handle multiple camera feeds, or just single video sources?
Yes, multi-camera video analytics setup is included, syncing and processing feeds from several cameras into one unified system.
Will you test the model's accuracy before delivery?
Yes, accuracy testing is done before delivery, verified against real data rather than shipped untested, with results like 92% face emotion detection accuracy.
Can you extract text from images, not just detect objects?
Yes, OCR and text extraction from images are included, alongside object detection, image recognition, and segmentation work.
