• 8 min read
How to Add an LLM to a Vision Pipeline (And When to Avoid It)
Discover when to add an LLM to your computer vision pipeline. Learn how to combine object detection and visual reasoning in Roboflow Workflows.
• 8 min read
OCR Lot Code and Expiry Date Verification for Medical Packaging
Train a Roboflow localization model, isolate printed label fields, and verify batch numbers and expiry dates with Google Gemini.
• 12 min read
Give My Agent Eyes
Combine RF-DETR and LMMs to build an AI pipeline that perceives, reasons, and acts.
• 7 min read
Turn Vision AI Detections Into Alerts: PLC, MES, Slack
Learn how to detect defects with Roboflow and trigger real-time downstream alerts in manufacturing facilities.
• 15 min read
PPE Detection with AI: Real-Time Worker Safety Monitoring
Learn how to automate workplace safety monitoring. Build a real-time PPE detection pipeline using RF-DETR and Roboflow Workflows to track helmets and vests.
• 8 min read
Pothole Detection with RF-DETR
Discover how to build a pothole detection pipeline. Use RF-DETR and ByteTrack in Roboflow Workflows to automate road hazard tracking and repair prioritization.
• 7 min read
Tracking Baseball Games with AI: Dynamic Field Analytics
Build a high-speed computer vision pipeline using RF-DETR and OC-SORT to extract player and ball telemetry data.
• 27 min read
Automate Food Service with AI
Two tutorials for automating food quality with computer vision: end-line plating QA with Gemini 3, and a packaging line monitor with RF-DETR and ByteTrack.
• 7 min read
High-Speed Analytics: Tracking Table Tennis Movements with AI
Build a high-speed computer vision pipeline using RF-DETR and OC-SORT to generate sports analytics.
• 6 min read
Pickleball Player Performance Analytics with Roboflow
Analyze pickleball player positioning and tactics automatically using RF-DETR, Roboflow Workflows, and Claude.
• 7 min read
Hockey Player Tracking with Roboflow
Learn how to build a hockey tracking pipeline. Use RF-DETR and ByteTrack in Roboflow Workflows to track players and visualize movement trails.
• 6 min read
Automated Volleyball Tracking with RF-DETR and ByteTracker
Learn how to build a real-time volleyball tracking pipeline. Combine RF-DETR and ByteTrack in Roboflow Workflows for automated sports analytics.
• 7 min read
Tennis Player Performance Analytics with Roboflow
Build an automated tennis player positioning analysis system using RF-DETR, Roboflow Train, and Workflows.
• 7 min read
Vision Token Counts: What does it cost to process an image with a frontier vision model?
Understand the cost, per-provider tokenization rules, and a comparison across image sizes for Claude, GPT, and Gemini.
• 5 min read
AI Swimmer Tracking
Learn how to build an automated aquatic monitoring pipeline using RF-DETR and ByteTrack in Roboflow.
• 4 min read
Latency vs Accuracy Tradeoff in Object Detection, Solved
Solve the latency vs accuracy tradeoff in object detection. Watch NAS search thousands of models and map the curve.
• 17 min read
Detecting Road Defects with Computer Vision
This tutorial shows how to build an automated road defect detection system using Roboflow with deployment strategies for UAV imagery and dashcam video, each generating structured inspection reports automatically.
• 8 min read
AI Crop Analysis with Roboflow
Learn how to combine RF-DETR and Claude Sonnet 4.5 in Roboflow Workflows to detect plant diseases.
• 9 min read
Trading Card Inventory Management with AI
Learn how to build an automated TCG inventory pipeline using RF-DETR and Gemini 2.5 Flash to extract card details and market values.
• 5 min read
Serverless GPU Inference Cost Comparison: Roboflow, GCP, AWS, Azure
SUMMARY
Serving a custom RF-DETR XL model on serverless GPU infrastructure produces dramatically different monthly costs depending on the provider and traffic pattern, so this post benchmarks Roboflow Serverless, GCP Cloud Run, AWS SageMaker, and Azure Serverless GPU across three workloads: continuous inference at one request per 10 seconds,
• 21 min read
DINOv3: An Advanced Self-Supervised Vision Foundation Model by Meta
Explore Meta's DINOv3, the state-of-the-art self-supervised vision foundation model. Learn how its SSL architecture works and how to train it on Roboflow.
• 11 min read
Automate Surgical Instrument Tracking with Computer Vision
See how to build a surgical instrument closing-count assistant with Roboflow.
• 8 min read
High-Precision OCR for Medical Device Labeling with RF-DETR and Gemini 2.5 Flash
Learn how to build a two-stage Vision Agent using RF-DETR and Gemini 2.5 Flash for medical device labeling.
• 7 min read
Turning Any Screen Into a Touchless Display with Egocentric Hand Keypoint Detection
In this guide, we’re moving beyond the glass to build a first-person interaction system. Using Roboflow Inference 1.0, RF-DETR Nano, and MediaPipe, you’ll learn how to anchor a gesture space to a physical screen and translate hand landmarks into OS-level commands like zooming and scrolling.
Stay Connected