• 9 min read
How Often Should You Retrain a Computer Vision Model?
Learn how to recognize performance decline, data drift, new edge cases, and other signs that indicate a computer vision model needs retraining.
• 18 min read
Human-Object Interaction Detection with RF-DETR
Build human-object interaction detection without interaction labels: RF-DETR finds people and forklifts, Gemini classifies safe vs unsafe, in Roboflow.
• 14 min read
Computer Vision PLC Integration: Turn Detections into Machine Actions
Computer vision PLC integration turns detections into real machine actions. Learn how cameras, RF-DETR, Roboflow Workflows, edge inference, and protocols such as OPC UA, Modbus TCP, and EtherNet/IP connect vision systems to PLCs for reject, stop, alert, and quality-control tasks.
• 9 min read
Powder-Coat Defect Detection
Detect craters, orange peel, bubbles, and scratches with RF-DETR, then use Gemini 2.5 Pro to summarize the visible defect and its apparent severity in the annotated image.
• 5 min read
Building a Pool Monitor with Raspberry Pi and Roboflow
How to use a Raspberry Pi, a phone camera, and a Roboflow classification model to know when a pool needs a refill.
• 10 min read
Machine-Vision AI for Surface Defect Severity Assessment
Build a computer vision pipeline for identifying surface defect severity and evaluating product quality.
• 23 min read
Shop Floor Data Collection: How to Automate It with Computer Vision
Shop floor data is usually typed into forms after the fact. Cameras already see every unit, defect, and stoppage. Learn how to turn those observations into timestamped production events with RF-DETR, Roboflow Workflows, and Vision Events, including a working cycle-time example.
• 9 min read
What Is a Unified Namespace? Where Vision AI Fits in the UNS
A unified namespace gives every plant system one data model. Roboflow Workflows publish camera detections as MQTT events any MES can read.
• 10 min read
Retail Object Detection with RF-DETR
Train a model to detect empty shelf spaces and automate restocking decisions.
• 22 min read
Production Line Monitoring With Camera AI
Build production line monitoring with camera AI: train RF-DETR to detect defects, count bottles, and trigger alerts with Roboflow Workflows.
• 13 min read
How to Make Automatic Highlight Reels from Kids' Soccer Games
Upload a soccer game, click your kid once, and get back a highlight reel with every play they were part of. Two RF-DETR models trained in the Roboflow UI handle detection and tracking, and a local Python pipeline reads jersey numbers, finds the goals, and cuts the clips.
• 6 min read
Run RF-DETR in NVIDIA DeepStream on Jetson
From pretrained weights to live multi-camera inference on a Jetson Orin NX: TensorRT engine build, the custom bbox parser DeepStream needs, and per-class colors with a pyds probe.
• 8 min read
Hog Ring Detection with Computer Vision
Detect visible hog rings with RF-DETR, compare their positions with predefined attachment zones, and use Gemini 2.5 Pro to generate an automotive seat inspection summary.
• 4 min read
Gemini 3.6 Flash for Vision: Evaluation and Benchmarks
Cheaper and faster than Gemini 3.5 Flash, and the best model we tested on video. Object detection is the one real step back.
• 10 min read
Flanges Quality Inspection with Computer Vision
Train RF-DETR to catch flange face defects, then sort results into pass, review, and fail.
• 13 min read
Advanced Techniques for Optimizing AI Inference Costs
In this guide, you will learn about advanced techniques that can help reduce AI inference costs across the complete computer vision pipeline.
• 9 min read
Pipe and Tubes Quality Inspection with Roboflow
Automate tube and pipe defect inspection. Train an RF-DETR model to detect holes, cracks, and ruptures, and build a pass, review, or fail Workflow in Roboflow.
• 9 min read
Retail Object Detection with RF-DETR
Detect visible beverage products with RF-DETR, compare their counts with fixed camera-frame thresholds, and use Gemini 2.5 Pro to generate a shelf inspection summary.
• 17 min read
Teaching a Porch to Recognize Delivery Drivers and Accept Packages
An end-to-end build of a self-unlocking porch lockbox: on-device detection on an old iPhone gates a cloud Roboflow Workflow, which validates courier-with-package events inside a drawn porch zone and triggers a solenoid lock through an ESP32. Includes delivery verification and a live dashboard.
• 11 min read
Cosmetic Defect Detection with Computer Vision
Learn how computer vision can automate cosmetic defect inspection by detecting paint scratches and identifying the affected car part. This guide shows how to train an RF-DETR model in Roboflow, build a part-level inspection Workflow, and return visual results and repair-ready outputs for QA review.
• 6 min read
Multi-Model Auto Labeling for Segmentation with Roboflow Workflows
Stack SAM 3, Gemini, and GPT into a single Workflow, require a 2-of-3 vote on every mask, and auto-label entire segmentation datasets without drawing a single polygon.
• 6 min read
GPT 5.6 Sol is the best "vision" model OpenAI ever released
GPT-5.6 Sol is OpenAI’s strongest vision model yet. We tested Sol, Terra, and Luna across detection, counting, OCR, and extraction, then compared their results, speed, and cost with leading VLMs.
• 4 min read
How to Scale a Computer Vision Pilot to Production in Manufacturing
See how to scale a computer vision pilot to production: why 77% stall on integration, not the model, and how to get past it in manufacturing.
• 21 min read
Best Computer Vision Models in 2026: A Task-by-Task Guide
Explore the best computer vision models in 2026 for object detection, segmentation, classification, keypoints, OCR, vision-language tasks, depth estimation, and tracking. Compare leading models, benchmarks, licensing, deployment options, and when to fine-tune on custom data using Roboflow tools.
Stay Connected