We build computer vision systems that see, read, and decide.
Rizwan AI builds computer vision and deep learning systems that turn images and video into reliable, automated decisions. Built to run in production on cloud, on-prem, or edge devices.
Free computer vision tools and ONNX model analyzer
Three things you can use in the next five minutes: a browser-based ONNX viewer and model inspector that visualizes the network graph and analyzes it for problems, a dataset linter that audits YOLO and COCO datasets before they waste a training run, and an on-device video redaction app for macOS. Each one runs on your own machine, and none of them asks for an account.
- Browser tool
ONNX Visualizer & Model Inspector
View, analyze and inspect ONNX models with architecture diagrams
- Health, optimisation and runtime-compatibility analysis
- Visualize the neural network graph; benchmark and diff models
- Export architecture diagrams as PNG, SVG, AVIF, JPG or animation
Any modern browserNothing to installOpen → - Desktop appNew
Redact
Hide faces and objects in video, on your device
- On-device YOLO26 segmentation; nothing is uploaded
- A live webcam feed, or a saved video or image
- Box or mask, pixelate or blur, across 80 classes
macOSv1.0.0Open → - Command lineNew
CVFlow
Audit a detection dataset before it wastes a training run
- 27 checks: corrupt files, bad boxes, duplicates, leakage
- Terminal report, browser dashboard or one HTML file
- Fix YOLO labels in place; --strict gates a CI build
Windows, macOS & Linuxv0.0.3Open →
Computer vision, built for production
Every engagement runs the same way: the problem framed against your own data before anything is built, a prototype early enough that you can still change your mind, then hardening into something that holds up under real traffic.
Object detection
Find and count objects in real time
Object tracking
Follow people and vehicles across frames
Instance segmentation
Pixel-accurate masks around every object
Image classification
Sort images by category or defect
Pose estimation
Track body keypoints and posture
Depth estimation
Judge distance from a single camera
OCR & document AI
Read text, fields and tables
Video analytics
Counts, dwell time and zone alerts
Quality inspection
Catch defects on the production line
Anomaly detection
Flag rare and abnormal events
Data labelling & audit
Curate, label and clean training data
Model training & tuning
Train, evaluate and lift accuracy
Edge deployment
Run on Jetson, mobile and embedded
Cloud & on-prem deployment
Ship to your servers or cloud
Monitoring & retraining
Watch drift, retrain, hold accuracy
What clients say
Muhammad Rizwan Munawar has delivered computer vision and deep learning projects for clients worldwide. Here is what a few of them said about working together. Every review is verified on Upwork.
6 🌟 reviews on- $2,620.80
Excellent to work with and exceeded all expectations in answering questions related to various courses. One of the standout qualities of Muhammed was their commitment to going above and beyond. They didn't just provide brief, surface-level answers. Instead, they took the time to provide in-depth responses that demonstrated their expertise and dedication to the job
VerifiedAugmented AIComputer vision project - $1700 / 55$/hr
Muhammad did an excellent job on my project. He's very knowledgeable, intelligent, and highly capable. I hope to work with Muhammad again in the future.
VerifiedJoe GratzDeep learning project - $600
Even though he had to work till midnight, Rizwan always worked diligently and attempted to solve difficulties. Rizwan is the best freelancer with whom I've ever worked in Upwork.
VerifiedYanuar BramantyoComputer vision project - $317.50 / 35$/hr
Good experience, was available when needed and on time on our appointment. explained well and teached well.
VerifiedDarmandran SivapathasundramComputer vision project - $270
Work was completed promptly and with great care. He's willing to go the extra mile and was willing to adapt to changes. Great job!
VerifiedAnn DelaneyComputer vision project - $400
Muhammad was very professional, helpful, and always available throughout the project. He provided valuable support and contributed significantly to the development of our solution. It was a great experience working with him, and I would definitely recommend him for similar projects. I hope we have the opportunity to work together again in the future.
VerifiedAndre VasquesComputer vision project
Hands-On AI: Computer Vision Projects with Ultralytics and OpenCV
Build real computer vision projects from scratch using Ultralytics YOLO and OpenCV.
- Object counting in images and live video
- Queue management and waiting-line analysis
- Track and count objects inside custom zones
- Turn detections into analytical graphs and charts
- Run real-time YOLO inference in a Streamlit web app
From the blog
Tutorials, code, and notes on computer vision, deep learning, and applied AI.
Depth EstimationJuly 27, 20269 min readVideo Depth Anything in Python: consistent depth for video
Depth Anything V2 run frame by frame flickers on video. Video Depth Anything fixes that with temporally consistent depth for long clips. Here is how to run it in Python, with the command line and a reusable script.
YOLO26July 19, 202619 min readYOLO26 vs YOLO11 vs YOLOv8: which one should you use?
Six generations of YOLO side by side: published accuracy and parameter counts for YOLOv5, YOLOv8, YOLOv9, YOLOv10, YOLO11 and YOLO26, what actually changed between them, and a straight answer on which one to use for detection, edge deployment, or transfer learning.
OCRJuly 18, 20267 min readHow to extract text from images in Python (OCR): 5 libraries compared
A practical guide to reading text from images in Python: Tesseract, EasyOCR, PaddleOCR, docTR and VLM-based OCR, with runnable code and a which-one-to-pick guide.
Object TrackingJuly 18, 20268 min readHow to use ByteTrack with YOLO for object tracking in Python
Give YOLO a memory. This tutorial uses ByteTrack to assign persistent IDs to objects across video frames, with a runnable Ultralytics script and a bytetrack.yaml tuning guide.
Semantic SearchJuly 6, 20267 min readBuild a semantic image search engine with CLIP and Python
Learn how to build a semantic image search engine that finds pictures by meaning. A few lines of Python turn a folder of images into a searchable index you can query in plain English.
YOLO26July 3, 20269 min readYOLO26 vs YOLO11: Real-time ONNX FPS benchmark in Python
Build one small, reusable class that runs Ultralytics YOLO26 and Ultralytics YOLO11 as ONNX models, draws clean detections, and overlays live FPS and latency so you can compare their real-time speed on the exact same footage.
Depth EstimationJuly 1, 202615 min readDepth Anything V2 in Python: depth from image, video or webcam
How to estimate depth from a single image, a video, or your webcam with one clean, reusable Python class, and how V2 compares to V1 and Depth Anything 3.
Object TrackingJune 27, 202613 min readUltralytics object trackers comparison: ByteTrack, BoT-SORT & More
How do the six Ultralytics trackers actually behave on the same footage? A look at BoT-SORT, ByteTrack, OC-SORT, Deep OC-SORT, FastTrack, and TrackTrack, their internals, trade-offs, and side-by-side results on ID switches, ID stability, and FPS.