You will get Computer-Vision People Counting projects
Project details
You will get a AI Computer Vision solution designed to meet your specific needs. I specialize in developing high-accuracy models for Object Detection, Segmentation, and Oriented Bounding Boxes using state-of-the-art technologies like YOLO.
With real-time inference and 95%+ precision, these models are ideal for applications in industries like logistics, retail, robotics, and healthcare. My extensive experience in TensorFlow, PyTorch, and OpenCV ensures seamless integration and deployment for your business.
Whether you provide an annotated dataset or need end-to-end support, I will deliver a solution optimized for speed, accuracy, and scalability.
With real-time inference and 95%+ precision, these models are ideal for applications in industries like logistics, retail, robotics, and healthcare. My extensive experience in TensorFlow, PyTorch, and OpenCV ensures seamless integration and deployment for your business.
Whether you provide an annotated dataset or need end-to-end support, I will deliver a solution optimized for speed, accuracy, and scalability.
AI Development Type
Deep Learning, Model Tuning, Software MaintenanceAI Tools
Azure Machine Learning, Keras, NVIDIA AI Platform, OpenCV, PyTorch, TensorFlowAI Development Language
PythonWhat's included
| Service Tiers |
Starter
$150
|
Standard
$200
|
Advanced
$700
|
|---|---|---|---|
| Delivery Time | 2 days | 5 days | 15 days |
Number of Revisions | 1 | 2 | 3 |
AI Model Integration | |||
Detailed Code Comments | |||
Knowledge Graph | - | - | |
Model Documentation | - | ||
Ontology | - | - | |
Source Code | |||
Taxonomy | - | - |
Frequently asked questions
5 reviews
(4)
(0)
(0)
(0)
(1)
This project doesn't have any reviews.
MD
Matthew D.
Apr 14, 2026
AI Developer for Wildlife Image Analysis (Detection, Classification, Matching)
Great Perfessional freelancer. Will work again together for future projects, Rizwan has exceeded my expectations and I have worked with many freelances on Upwork, I can safely say he is one of the best.
AK
Anthony K.
Oct 12, 2025
Facial Recognition
Nice guy and started well. However, it has been four months, and it's still incomplete and not functioning.
The bare minimum was done in some parts, other parts weren't done, or didn't follow the instructions provided (for example, using different software than requested). We did request minor tweaks, but that wasn't done either.
Upwork automatically dispersed the payment - I wouldn't have approved it otherwise. He then became extremely unresponsive (despite promising otherwise) to me and the team, and 3rd party auditors trying to review his work. Kept saying he still needed to do parts but never did. I had to keep aggressively following up, asking for updates over a month since then, and only got a couple of messages saying he will answer. He said he'd refund me but still hasn't.
The documentation provided was extremely clear. What was meant to take 1.5 weeks took 4+ months and still not functioning. It held back the rest of the team and the entire app production... for 4 months. I know need to find someone else to do it and will have to pay more to rush it.
I'm reaching the end of the 1 month window Upwork offers to mediate (and I told him this) so I have no choice but to end the contract and leave a review reflecting my experience. I got upwork involved to mediate, ending the contract to try to find a resolution.
The bare minimum was done in some parts, other parts weren't done, or didn't follow the instructions provided (for example, using different software than requested). We did request minor tweaks, but that wasn't done either.
Upwork automatically dispersed the payment - I wouldn't have approved it otherwise. He then became extremely unresponsive (despite promising otherwise) to me and the team, and 3rd party auditors trying to review his work. Kept saying he still needed to do parts but never did. I had to keep aggressively following up, asking for updates over a month since then, and only got a couple of messages saying he will answer. He said he'd refund me but still hasn't.
The documentation provided was extremely clear. What was meant to take 1.5 weeks took 4+ months and still not functioning. It held back the rest of the team and the entire app production... for 4 months. I know need to find someone else to do it and will have to pay more to rush it.
I'm reaching the end of the 1 month window Upwork offers to mediate (and I told him this) so I have no choice but to end the contract and leave a review reflecting my experience. I got upwork involved to mediate, ending the contract to try to find a resolution.
RA
Rania A.
Oct 29, 2024
Academic: Orange Data Mining Software Experience needed... Data Analysis and Applications
MS
Michael Kofi Seer S.
Oct 20, 2024
AI Development for Brain Cancer Detection Application
JM
Jirou M.
Jan 31, 2024
Data Annotation for Images
Rizwan has been pretty helpful in his work. He is very reliable and he always works on time/deadline.
About Rizwan
Full Stack Machine Learning | Computer Vision Engineer| Edge AI | YOLO
100%
Job Success
Gujranwala, Pakistan - 10:46 am local time
If you're dealing with missed defects, inaccurate counting, or unreliable surveillance systems, I can design and deploy AI-powered solutions that run efficiently on edge devices (Jetson, Raspberry Pi) or in the cloud.
WHAT I BUILD FOR YOU
✅ Object Detection & Multi-Object Tracking (YOLOv5, YOLOv8, ByteTrack, DeepSORT, BOTSort)
✅ Real-Time Video Analytics — people counting, crowd monitoring, heatmaps
✅ Industrial Inspection & Defect Detection (manufacturing QA automation)
✅ Smart Surveillance Systems — intrusion detection, zone monitoring, alerts
✅ OCR & Document AI — license plates, barcodes, form data extraction
✅ Face Recognition & Identity Verification Systems
✅ Custom Computer Vision Models — training, fine-tuning, evaluation
EDGE AI & DEPLOYMENT (MY CORE STRENGTH)
✅ NVIDIA Jetson (Nano, Orin), Raspberry Pi, Hailo deployment
✅ TensorRT, ONNX optimization (low latency, high FPS)
✅ OpenCV real-time pipelines & video processing
✅ FastAPI / REST APIs / Docker-based deployment
✅ Cloud + GPU inference APIs (AWS-ready systems)
I don’t just train models, I build complete, production-ready systems optimized for real-world constraints.
REAL-WORLD SOLUTIONS I’VE BUILT
✔️ Real-time fall detection system on NVIDIA Jetson
✔️ Industrial defect detection pipelines for manufacturing
✔️ Zone monitoring & object tracking systems
✔️ Optimized inference pipelines using TensorRT + ONNX
INDUSTRIES I WORK WITH
Retail Analytics | Industrial QA | Smart Security | Healthcare Imaging
Logistics & Warehousing | Agriculture | Smart Cities | Surveillance
Wild life and Geo images analysis
HOW I WORK
1. Understand your use case & define success metrics
2. Design system architecture (model + pipeline + deployment)
3. Build, optimize, and test with real-world data
4. Deliver scalable, production-ready solutions
WHY CLIENTS CHOOSE ME
✔️ Production-focused (not just demos)
✔️ High-performance systems (low latency, optimized FPS)
✔️ Clean, maintainable, scalable code
✔️ Clear communication & reliable delivery
✔️ Long-term support after deployment
If you need a computer vision engineer who understands both AI models and real-world deployment challenges, let’s talk.
📩 Message me with your use case, and I’ll suggest the best approach for your project.
Steps for completing your project
After purchasing the project, send requirements so Rizwan can start the project.
Delivery time starts when Rizwan receives requirements from you.
Rizwan works on your project following the steps below.
Revisions may occur after the delivery date.
Requirement Analysis & Goal Alignment
Kickoff Meeting: Discuss your use case, target metrics (accuracy, speed), and constraints (hardware, compliance). Define Success Criteria: Agree on KPIs (e.g., ≥95% mAP, 30 FPS) and final deliverables (e.g., ONNX model, API, or embedded code).
Data Preparation & Annotation
Review data for quality, format, and annotation consistency. Use tools like CVAT or Label Studio for labeling. Augment with rotations/scaling, and split into train, validation, and test sets.