I'm Muhammad Rizwan Munawar, lead computer vision engineer at Rizwan AI and a content creator working with Ultralytics. I build real-time systems for object detection, segmentation, tracking, and pose estimation, and I share open-source tools, tutorials, and research with a large developer community.
Today my focus is computer vision content creation and consulting. I write tutorials and technical articles, publish open-source projects, mentor developers, and take on freelance vision work through Rizwan AI and Upwork.
I've been an active contributor to ultralytics/ultralytics, where I helped build much of the Solutions module, production-ready building blocks used by developers worldwide, along with docs, notebooks, and tutorials from YOLOv8 through YOLO11.
Teams and companies I've worked with or built projects for across the computer vision industry:
Trajectory Forecasting
trajectory-forecasting-demo-github.mp4
Predicts where tracked objects move next using Kalman-filtered forecasting on Ultralytics YOLO. View repository
SAM 3 Inference
sam3.1-demo-github.mp4
Promptable detection, segmentation, and auto-annotation with Meta Segment Anything Model 3. View repository
BlurIt
blur-it-v1.0-demo.mp4
Privacy-first image redaction in the browser. Click a face, download a clean PNG. View repository
CVFlow
cvflow-demo-github.mp4
CLI linter that audits YOLO and COCO datasets for broken images, bad labels, duplicates, and split leakage. View repository
| 🚀 Project | ⭐ Stars | 📚 Forks | 🛎 Issues | 📬 Pull requests |
| cvflow: CLI linter that audits YOLO/COCO datasets for broken images, bad labels, duplicates, and split leakage |  |
 |
 |
 |
| sima-projects: Reference YOLO app + setup guide for the SiMa.ai Modalix edge DevKit |  |
 |
 |
 |
| blurit: Privacy-first, in-browser image redaction; click a face, download a clean PNG |  |
 |
 |
 |
| trajectory-forcast: Kalman-filtered trajectory forecasting for Ultralytics YOLO |  |
 |
 |
 |
| streamgrid: Multi-stream video inference with Ultralytics YOLO in a grid layout |  |
 |
 |
 |
| sam3-inference: Inference toolkit for Meta Segment Anything Model 3 (SAM 3) |  |
 |
 |
 |
| ⭐ Project | ⭐ Stars | 📚 Forks | 🛎 Issues | 📬 Pull requests |
| yolov7-object-tracking: Real-time YOLOv7 object tracking with SORT |  |
 |
 |
 |
| yolov7-pose-estimation: Human pose and keypoint estimation with YOLOv7 |  |
 |
 |
 |
| yolov8-object-tracking: YOLOv8 object tracking with PyTorch, OpenCV, and Ultralytics |  |
 |
 |
 |
| yolov7-segmentation: Instance segmentation with YOLOv7 |  |
 |
 |
 |
| yolov5-object-tracking: YOLOv5 detection, tracking, blurring, and a Streamlit dashboard |  |
 |
 |
 |
| yolov7-object-blurring: Privacy-preserving object blurring with YOLOv7 |  |
 |
 |
 |
- 🔥 YOLO11 Object Detection and Instance Segmentation
A hands-on guide to detection and segmentation with Ultralytics YOLO11. - 🔥 TrackZone: Object Tracking in Regions using YOLO11
Track objects only inside the regions that matter for faster, cleaner analytics. - Smart Parking Management with YOLO11
Detect free and occupied parking slots in real time. - Workout Monitoring using YOLO11
Count reps and track form using pose estimation. - Monetizing Computer Vision Hobby Projects
How small vision projects turned into real earnings. - Roadmap for a Computer Vision Engineer
The path and resources to break into computer vision. - Becoming a Computer Vision Engineer (Ultralytics Blog)
Lessons from the field, featured on the Ultralytics blog.
- Automated Pallet Racking Inspection, Sensors (MDPI), 2022
- Diabetic Retinopathy Exudate Detection, IEEE, 2022
- Rice Leaf Defect Detection, Foods (MDPI), 2022
- Comparative Study of YOLO Models, CS & IT (AIRCC), 2022
- Explainable AI for Drug Sensitivity, IEEE, 2022
- More on Google Scholar
|
|
|
