CERBERUS: Crack Evaluation & Recognition Benchmark for Engineering Reliability & Urban Stability
This work addresses the need for reliable defect detection in infrastructure inspection, but it is incremental as it builds on existing methods like YOLO with a new benchmark.
The authors tackled the problem of training and evaluating AI models for detecting cracks in infrastructure by creating CERBERUS, a synthetic benchmark with a crack image generator and realistic 3D inspection scenarios. Results show that combining synthetic and real data improves performance on real-world images, though specific numerical gains are not provided.
CERBERUS is a synthetic benchmark designed to help train and evaluate AI models for detecting cracks and other defects in infrastructure. It includes a crack image generator and realistic 3D inspection scenarios built in Unity. The benchmark features two types of setups: a simple Fly-By wall inspection and a more complex Underpass scene with lighting and geometry challenges. We tested a popular object detection model (YOLO) using different combinations of synthetic and real crack data. Results show that combining synthetic and real data improves performance on real-world images. CERBERUS provides a flexible, repeatable way to test defect detection systems and supports future research in automated infrastructure inspection. CERBERUS is publicly available at https://github.com/justinreinman/Cerberus-Defect-Generator.