ROADWork Dataset: Learning to Recognize, Observe, Analyze and Drive Through Work Zones

Author/Presenter: Ghosh, Anurag; Tamburo, Robert; Zheng, Shen; Alvarez-Padilla, Juan R.; Zhu, Hailiang; Cardei, Michael; Dunn, Nicholas; Mertz, Christoph; Narasimhan, Srinivasa G.
Abstract:

Perceiving and navigating through work zones is challenging and under-explored, even with major strides in self-driving research. An important reason is the lack of open datasets for developing new algorithms to address this long-tailed scenario. The researchers propose the ROADWork dataset to learn how to recognize, observe and analyze and drive through work zones. They find that state-of-the-art foundation models perform poorly on work zones. With their dataset, they improve upon detecting work zone objects (+26.2 AP), while discovering work zones with higher precision (+32.5%) at a much higher discovery rate (12.8 times), significantly improve detecting (+23.9 AP) and reading (+14.2% 1-NED) work zone signs and describing work zones (+36.7 SPICE). They also compute drivable paths from work zone navigation videos and show that it is possible to predict navigational goals and pathways such that 53.6% goals have angular error (AE) < 0.5 degrees (+9.9 %) and 75.3% pathways have AE < 0.5 degrees (+8.1 %).