Researchers Release Bangladesh Rickshaw Traffic Dataset for AI and Autonomous Driving

A team has published the Bangladesh Rickshaw Traffic Dataset, aimed at advancing Physical AI, computer vision, robotics, and autonomous driving research. The dataset captures complex urban traffic scenes in Dhaka, featuring rickshaws, motorcycles, buses, trucks, pedestrians, and street vendors interacting in unstructured road environments. Unlike most publicly available traffic datasets collected in predictable, structured settings, this dataset reflects the chaotic real-world conditions typical of South Asian cities. Each video clip is paired with structured metadata covering scene category, object statistics, quality indicators, and privacy processing details to streamline integration into machine learning workflows. Free sample data has been made available on GitHub, Hugging Face, and Kaggle, with future releases planned to include night driving, adverse weather conditions, and additional cities.