We built a two-stage computer-vision pipeline that first detects traffic signs with YOLO, then classifies the cropped regions with a CNN.
A potential foundation for driver-assistance and map-updating scenarios, this R&D work demonstrates our experience combining detection and classification models in a single production line.