🌟 Editor's Note

Welcome to another exciting week in the Vision AI ecosystem! We've got a packed newsletter full of insights, events, and inspiring stories from the heart of innovation.

🗓️ Event Highlights

Physical AI at Automate 2025:

A recap of Automate 2025 highlights a significant leap forward in machine vision, moving from an emerging technology to a central driver of automation. The key trend was the shift toward tangible Physical AI, powered by affordable, embedded processing. This allows robots to perform complex tasks and quality control systems to detect flaws in milliseconds, enabling real-time, intelligent automation.

This evolution puts immense pressure on edge computing, a challenge directly addressed by the new OnLogic Helix 520 Series. The Helix 520, launched at the event, is purpose-built for these advanced applications. With its integrated Intel Core Ultra NPU for AI acceleration and scalable GPU power, it handles complex machine vision workloads directly at the source. This rugged, reliable, and modular device is designed for the toughest factory floors, solidifying OnLogic’s position in the future of automation. [link]

Deadlines:

• 3DV’2026 : 12 days

• WACV’26 (R2, reg) : 38 days

• ICLR’26 (abs) : 45 days

🚀 Case study

The Rise of Vision AI in battery manufacturing

Computer vision is reinventing battery manufacturing, moving beyond manual inspections. The technology enhances quality and speed by detecting defects like bubbles on electrode surfaces, ensuring precise layer stacking, and verifying packaging. Despite challenges like data security, this AI integration is crucial for meeting the surging demand for batteries. [link]

🦄 Startup Spotlight

Shelfmark : AI- powered visual inspection for manufacturing

Shelfmark, a startup founded by a CMU alumnus, has secured an investment from TitletownTech, a venture capital firm backed by Microsoft and the Green Bay Packers. The company uses AI-powered computer vision to provide manufacturers with automated visual inspection, helping to reduce defects and labor costs. Shelfmark's platform also focuses on inventory management for the Direct Store Delivery (DSD) market, providing real-time data to help suppliers and retailers optimize operations. [link]

🔥 Paper to Factory

Coarse-to-fine crack cue for robust crack detection

Traditional deep learning methods for crack detection often fail to generalize to new environments because they overlook the thin structure of cracks. CrackCue is a new method that addresses this by using a coarse-to-fine process to generate a robust "crack cue." This cue, which is unaffected by shadows or complex backgrounds, helps guide detection. As a plug-and-play solution, CrackCue can be integrated into existing networks, significantly boosting their generalization ability and robustness.

🏆 Community Spotlight:

• In the recent Voxel 51 podcast, Jason Corso highlights the rise of Visual AI applications in manufacturing

• In the latest Hansel minutes podcast, Sky Engine AI’s Dr. Marc Scouter dives deep into the surprising power of synthetic data, exploring when fake can outperform real in areas like medical imaging, defense and self driving cars

• In their latest Visual AI in manufacturing report, Voxel 51 discusses the future of AI in manufacturing, the leading companies and voices driving innovation in the space

• Roboflow’s latest tutorial highlights on how to annotate a dataset and train a vision AI model

Reddit / X corner:

• Ultralytics discuss on how to detect rotated objects with Ultralytics YOLO OBB task and how to train YOLO11 on the LVIS dataset for long-tail object detection

• A latest Reddit post discusses on problems in training EfficientDet model for Edge TPU

• Another reddit post discusses on training a segmentation model when an object has optional parts and annotations are inconsistent

Till next time,

Vision AI weekly