Vehicle Perception
Autonomous Driving Case Study
Computer Vision
“One of our priorities at DeGould is to supply our automotive customers with high accuracy models for vehicle inspection. Quality annotation significantly contributes to this service. As a company, we have been collaborating with Ingedata to ensure we achieve the quality annotation our customers require; Ingedata has proved to be a reliable partner that performs consistently over time.”
Dean Gould, Machine Learning Engineer @DeGould
Our applications in autonomous driving include image and Point Cloud annotation. This requires Ingedata to annotate video feeds, which are more complex to handle than images. Our annotation processes embed “tricks” that make annotating highly productive despite the large number of frames in a video stream.
The mission
Cameras are a crucial sensor for autonomous vehicles. EasyMile requested extremely precise segmentation and tagging to make the most of the data collected.
The outcome
Ingedata defined precise segmentation by adapting recommendations from Cityscapes to EasyMile’s specific needs and set up a specific team of annotators for pixel-precise segmentation.
“A baby learns to crawl, walk and then run. We are in the crawling stage when it comes to applying machine learning.”
Dave Water, Department of Earth Sciences, University of Oxford
Client stories
Our Case Studies
Learn more about our approach and explore our case studies