Challenge
Safety, efficiency, and reliability for railway operations are paramount. OTIV is a front-runner in rail technology with assistance, remote operations, and autonomous systems.
To succeed in their mission, OTIV needs accurate labeling of railways, traffic signs, and traffic lights.
They also need to identify and predict objects that can suddenly cross the rail, ranging from pedestrians to cars to trucks. For this, they need pixel-perfect annotations.
Solution
OTIV is using Segments.ai since 2021. The team integrates Segments.ai into their ML lifecycle, together with an external labeling team.
Multi-sensor labeling is used to increase the performance of ML models to detect persons and vehicles on and off the tracks.
Next to that, OTIV uses image segmentation, bounding boxes, and vector labeling to achieve world-class algorithms for their high-performance industrial-grade sensors.
OTIV at a glance
Benefits
Multi-sensor labeling
ML-powered image segmentation
Managing internal and external labeling teams
Segments.ai gives us the ability to easily share work internally and add new collaborators locally or remotely.
Sean Nachtrab
Software/ML Engineer
OTIV
