TMTom Mathewinridingbluewaves.hashnode.dev·6d ago · 11 min readRiding the Blue Wave: Building Autonomous Intelligence on the Edge #06I'll start with the confession. My README said the system ran at 10.3 FPS with a real-time parallel pipeline. When I sat down and actually measured it running with the display on, it managed 2–3 FPS, 00
TMTom Mathewinridingbluewaves.hashnode.dev·6d ago · 10 min readRiding the Blue Wave: Building Autonomous Intelligence on the Edge #05Every object detector gives you a confidence score, and the beginner's instinct is to trust it directly: high score means hazard, act on it. On a single image, fine. On a live video stream from a boun00
TMTom Mathewinridingbluewaves.hashnode.dev·6d ago · 12 min readRiding the Blue Wave: Building Autonomous Intelligence on the Edge #04Adding threads to a system is easy. Adding threads without turning it into a minefield of race conditions, priority inversions, and mysterious frame drops is the actual engineering. In Episode 2 I des00
TMTom Mathewinridingbluewaves.hashnode.dev·6d ago · 11 min readRiding the Blue Wave: Building Autonomous Intelligence on the Edge #03In the last episode I walked through the system architecture, the asynchronous multi-threaded pipeline that keeps the whole thing real-time. This episode goes one level deeper, into the part everyone 00
IEIyanuoluwa Enoch Okeinfreecodecamp.org·Jul 25 · 14 min readHow to Build a Real-Time Object Detection and Tracking Pipeline with ROS 2 and YOLOv11If you've ever tried to build a robotics system that can actually see, track, and respond to the world around it, you know that the hard part isn't training a detection model. The hard part is making 00
Ccircuitrocksincircuitrocks.hashnode.dev·Jul 19 · 2 min readRun YOLO Vision Models on Raspberry Pi 5 Using Intel OpenVINOA Raspberry Pi 5 sitting next to a USB camera can now run a YOLO object-detection model at usable frame rates, with no cloud GPU and no workstation in the loop. That is the setup Intel and Ultralytics00
MSManikanta SSBinssbb7.hashnode.dev·Jun 9 · 7 min readI Spent Weeks Fighting OCR Before Realizing I Was Solving the Wrong ProblemHeads-up: The figure captions in this article are clickable. Click on any figure caption to view the associated outputs, visualizations, and intermediate results discussed in that section. Hey guys, i00
CVChirag Vijayinchirag4862.hashnode.dev·Apr 29 · 8 min readI Fine-Tuned YOLO to Understand Document Structure — Here's How It WorksThere's a class of problem in document AI that sounds deceptively simple: look at a page, figure out what's on it. Not read the text. Not classify the document. Just answer: where is the table? where 00
SSOLARTODOinsolartodo.hashnode.dev·Apr 14 · 13 min readHow AI Traffic Systems Detect 30+ Violation Types in Under 50 Milliseconds — The Edge Computing Pipeline Behind Modern Enforcement Traditional traffic cameras take a photo when a loop detector senses a vehicle. The photo goes to a server. The server runs license plate recognition. A human operator reviews the image and confirms t00
OWOliver Wenginedge-insights.hashnode.dev·Apr 8 · 6 min read🚀 Edge AI & Efficient Computing | Bi-Weekly News: # 2🎉Welcome to the second post of Edge AI & Efficient Computing! 👋 I’m Yuqin (Oliver) Weng, and I’m incredibly passionate about pushing the boundaries of AI on constrained devices. 🧠I created this spa10