Я знаю, что ты делал этой ночью

Привет, Хабр! Сегодня с вами участники профессионального сообщества NTA Промкин Михаил, Мымрин Дмитрий и Господарикова Ирина. Одной из областей применения ИИ сегодня является автоматизация контроля за сотрудниками. В данном посте мы рассмотрим приложение технологий ML к задаче детектирования спящих людей (в частности, охранников на рабочем месте) по видеозаписям камер наблюдения. Обсудим технические аспекты этого процесса, а также потенциальные преимущества и перспективы, которые предоставляет применение искусственного интеллекта в обеспечении безопасности на рабочем месте. Узнать больше

https://habr.com/ru/articles/780262/

#обработка_видео #yolov7 #yolov7_pose #yolo #python

Я знаю, что ты делал этой ночью

Привет, Хабр! Сегодня с вами участники профессионального сообщества NTA Промкин Михаил, Мымрин Дмитрий и Господарикова Ирина. Одной из областей применения ИИ сегодня является автоматизация...

Хабр
Playing with #YOLOv7...

Hello #fosstodon,
In case you missed it, here's my post on how my team taught an #autonomous electric go-kart to recognize traffic cones. We used #machinelearning with the #YOLOv7 framework and about 11,000 labeled training images:

https://matthew.krupczak.org/2022/12/13/coneslayer-a-neural-network/
#PyTorch #Python #ML #computervision #racing #ev

Coneslayer - A lightweight neural network for rapid detection of traffic cones

Coneslayer: A lightweight neural-network for rapid detection of traffic cones. Based on YOLOv7, a state of the art object detection framework

Matt's Tech Bytes
Coneslayer - A lightweight neural network for detection of traffic cones

Introduction: Hello world! It is my pleasure to finally share the subject of my recent obsession: #machinelearning and traffic cones. Over the past several months, I've been working with undergraduate students of the Kennesaw State University Electric Vehicle (#racing) Team [ksuevt.

We made probably the best neural net in the world for detecting orange traffic cones that can run on a small edge device

Check it out here: https://github.com/mkrupczak3/Coneslayer

#ml #yolov7 #Luxonis
#oakd #rae #robotics

GitHub - mkrupczak3/Coneslayer: A lightweight neural-network for rapid detection of traffic cones

A lightweight neural-network for rapid detection of traffic cones - GitHub - mkrupczak3/Coneslayer: A lightweight neural-network for rapid detection of traffic cones

GitHub

Check out my latest article: Coneslayer - A lightweight neural network for detection of traffic cones https://www.linkedin.com/pulse/coneslayer-lightweight-neural-network-detection-traffic-krupczak via @LinkedIn

#machinelearning #AI #yolov7 #autonomousdriving #robotics #computervision

Coneslayer - A lightweight neural network for detection of traffic cones

Introduction: Hello world! It is my pleasure to finally share the subject of my recent obsession: #machinelearning and traffic cones. Over the past several months, I've been working with undergraduate students of the Kennesaw State University Electric Vehicle (#racing) Team [ksuevt.

Hello #Fosstodon,

I've just released "coneslayer" on behalf of the Kennesaw State University's Electric Vehicle Team. It is a lightweight neural network for detecting traffic cones based on a state of the art framework called #yolov7. It's designed to run live on edge devices at a framerate of at least 30fps.

https://github.com/mkrupczak3/Coneslayer

GitHub - mkrupczak3/Coneslayer: A lightweight neural-network for rapid detection of traffic cones

A lightweight neural-network for rapid detection of traffic cones - GitHub - mkrupczak3/Coneslayer: A lightweight neural-network for rapid detection of traffic cones

GitHub

Joys of working with an object detection model

I'm %62 sure that's a traffic cone 😂

#ml #yolov7 #objectdetection #machinelearning

The fruit of my team's labor, after countless hours of combined effort

There's a lot more to this story, I hope to do a write-up soon

#machinelearning #racing #yolo #cv #yolov7 #python #pytorch

Fixed, after training on this and 629 other hand-labeled images 😵‍💫:

Still have some false positives in other settings. I will add more negative training images to the training set

#machinelearning #yolov7 #ML #AI