Case Study: Chalmers University of Technology advances privacy-compliant micromobility research with brighter AI Precision Blur

Chalmers University of Technology anonymizes 8,000+ images with brighter AI

Chalmers University of Technology sought to create an open-source dataset for micromobility safety research but faced significant GDPR compliance risks. Capturing thousands of high-definition images in public urban spaces made obtaining individual consent impractical. The university partnered with brighter AI and its Precision Blur service to anonymize this sensitive data at scale while preserving its utility for training computer vision models.

Using brighter AI's automated detection and redaction technology, the team anonymized over 8,000 full-HD images. This solution enabled the successful release of the MicroVision dataset, which was used to train state-of-the-art object detection models that achieved a high mean average precision of up to 0.723. The project resulted in a publicly available, privacy-compliant dataset with over 30,000 annotations, providing a new benchmark for global AI-driven traffic safety research.

Chalmers University of Technology