Case Study: IntelinAir achieves pixel-accurate, high-quality annotations with SuperAnnotate

How IntelinAir built the highest quality datasets with SuperAnnotate

IntelinAir, an aerial imagery analytics company that uses computer vision and deep learning to support farmers, needed a better way to create pixel-accurate annotations for complex aerial imagery projects. The team had built an in-house tool and tried other platforms, but they were looking for a solution that could improve both annotation quality and speed. SuperAnnotate provided the platform they used to address this challenge.

With SuperAnnotate’s feature-rich annotation tooling, project management, and workflow capabilities, IntelinAir was able to produce more accurate training data and work more efficiently. The paintbrush functionality made difficult tasks like individual segmentation masks for corn-kernel counting feasible, and the team was able to upload data, assign work, QA quickly, and export to COCO format. IntelinAir said the results were higher-quality annotations and speeds previously unachievable with its prior tooling, with some projects becoming possible that were “not doable” before.

Intelinair

Jennifer Hobbs

Director of Machine Learning