Case Study: TACK Project speeds up image annotation with SuperAnnotate

How SuperAnnotate helps researchers in Sweden and Italy speed up image annotation

The TACK project, a research initiative conducted by universities in Sweden and Italy, faced the challenge of creating a large, accurately labeled dataset to train AI for detecting cracks in bridges and tunnels. Their manual image annotation process for their deep learning model was extremely time-consuming and inefficient. They adopted the SuperAnnotate platform to accelerate this crucial data preparation stage.

Using SuperAnnotate's tools, including its collaborative online platform and the "Smart Segmentation" feature, the research team significantly sped up the image annotation process. The smart tool partially auto-detected crack shapes, which greatly reduced the manual workload for annotators. This allowed the TACK project to efficiently label hundreds of images from their own and online datasets, which was vital for improving their model's performance and accuracy in real-world operational environments.

TACK Project