❌

Normal view

There are new articles available, click to refresh the page.
Before yesterdayMain stream

Announcing the OpenCV Perception Challenge for Bin-Picking

11 January 2025 at 02:35

We’re excited to announce a different type of competition than we’ve done in the past: The Perception Challenge For Bin-Picking, sponsored by Intrinsic, a new computer vision themed competition designed to tackle real-world robotics problems. This is your opportunity to compete and showcase your skills to a global audience, all while competing for a share of $60,000 in prizes! Read on to find out the rules and deadlines.

What is bin-picking? Bin picking is a robotics task that involves a camera and robotic arm or manipulator selecting individual objects from a container (often referred to as a bin) and moving them to a specific location or orientation. It is widely used in industries such as manufacturing, logistics, and warehousing to automate processes like sorting, assembly, and packaging.

Why Join?

This challenge is about putting your computer vision, AI, and robotics expertise to the test in real-world scenarios. Unlike previous OpenCV competitions, the goal here is well-defined, with your task being to create the most accurate results. Using Intrinsic’s open-source datasets and a remotely operated robotic workcell powered by the Intrinsic platform, you’ll develop cutting-edge solutions for challenging 6DoF pose estimation.

Unlike traditional competitions, we’re evaluating your algorithms being used live with a robot-in-the-loop, offering a more realistic and industrial type assessment of their success.

  • Don’t worry if you aren’t familiar with this type of vision task. We’ll explain more and provide example code in advance to people who register for the competition of the February 1st official opening for submissions. Don’t miss out: register today!

The Prizes

  • 1st (Grand Prize Winner): $18,000
  • 2nd (Grand Prize Runner-Up): $12,000
  • 3rd Place: $8,000
  • 4th Place: $4,000
  • 5th Place: $3,000
  • 6th to 10th Place: $1,000
  • Special Prize for Best One-shot Solution: $10,000

Key Dates

All deadlines are 11:59 pm (Pacific Time)

  • Today β€” Competition Registration Begins
  • February 1, 2025 β€” Competition Begins
  • May 1, 2025 β€” Top 5 Teams Announced (Elite Phase Begins)
  • May 22, 2025 β€” Live Bin-Picking Challenge on OpenCV Webinar
  • June 10, 2025 β€” Final Winners Awarded

How It Works

Participants will submit their solutions as Docker images for evaluation in test environments. Leaderboards will update in real-time via EvalAI, tracking accuracy and efficiency metrics.Β 

This challenge is open to teams of up to 10 members. Please note: participation is restricted to countries not subject to sanctions from the United States of America.

Watch The Announcement Webinar

Be Part of the Action

Looking for teammates to join the OpenCV Perception Challenge For Bin-Picking (sponsored by Intrinsic)? Connect with fellow participants on the official OpenCV Slack channel and OpenCV forum.

Whether you’re a seasoned professional or a new student in robotics and AI, this is your chance to make a mark. Join the competition, engage with the OpenCV community, and bring your ideas to life.

Visit bpc.opencv.org to learn more and register your team. The challenge runs in partnership with the BOP benchmark and we encourage participants to submit their results to the associated BOP-Robotics challenge, which will be announced soon.

The post Announcing the OpenCV Perception Challenge for Bin-Picking appeared first on OpenCV.

Seeed Studio at the 2nd UIC TrainRail Hackathon: Advancing Innovation in the Railway Industry

By: Ellie Cai
13 December 2024 at 09:17

From November 23rd to 24th, 2024, the 2nd UIC TrainRail Hackathon Asia-Pacific Regional Competition took place at Southwest Jiaotong University in China. The event attracted over 200 talented students from across the Asia-Pacific region, forming 42 teams to explore solutions for the future of the railway industry. With themes centered on enhancing passenger experience, strengthening freight capacity, and optimizing operational efficiency, the competition brought fresh perspectives and innovative technologies to the sector.

Committed to nurturing innovation and driving industry progress, Seeed Studio was thrilled to support the competition, together with other institutions such as Khon Kaen University in Thailand and Yamanashi University in Japan.

We were happy to provide comprehensive support throughout the competition, including pre-competition technical guidance and a wide range of hardware resources such as XIAO series, Grove sensors, Arduino kits, LoRa E5 modules, and Raspberry Pi modules, etc, empowering participants to rapidly prototype their ideas and overcome technical challenges.

Teams from Southwest Jiaotong University and Shijiazhuang Railway University claimed the top prizes for their exceptional projects. They will represent the Asia-Pacific region at the global finals in April 2025.

Highlights of Winning Projects

Participants showcased a range of innovative projects that directly addressed real-world railway challenges. Notable projects included:

  • Railway Wind and Snow Disaster Early Warning and Response System: This system provides real-time monitoring and early warnings for railway operations under adverse weather conditions, enhancing safety and efficiency.
  • Automated Barrier-Free Platform System: Designed to improve accessibility for passengers with disabilities, this intelligent platform solution optimizes barrier-free services in railway transportation.
  • Contact Network Arc Intelligent Monitoring System: Employing vision and sensor technologies, this system offers an intelligent solution for railway equipment maintenance, significantly improving efficiency and accuracy.
  • AI-Powered Real-time Rail Defect Detection: Using AI models to analyze rail conditions in real-time for enhanced safety monitoring.
  • Smoke Purification Device for Trains: Improving air quality within train cabins for a more comfortable passenger experience.
  • Vibration-Based Railway Construction Personnel Monitoring: Employing vibration wave technology to monitor the dynamic changes in construction environments and enhance worker safety.
  • Real-time Monitoring of Load Distribution in Freight Trains: Ensuring the smooth operation and safety of freight trains by monitoring load distribution during transit.
  • Computer Vision-Based Pantograph Modal Analysis: Optimizing pantograph monitoring through visual technology to improve the efficiency of electrified railway operations.

The hackathon underscores the vital connection between education, international collaboration, and industry development. Seeed Studio is proud to have been a part of this event and is committed to empowering young innovators to drive the digital transformation of the railway industry.

Congratulation to all participants for their hard work and creativity!

The post Seeed Studio at the 2nd UIC TrainRail Hackathon: Advancing Innovation in the Railway Industry appeared first on Latest Open Tech From Seeed.

❌
❌