Dual-modal Roadside Traffic is a dual-modal roadside traffic target detection dataset. It collects roadside traffic data through spatially calibrated and time-synchronized RGB cameras and event cameras, aiming to provide standardized data support for RGB and event multimodal fusion, event camera perception, and robust traffic scene understanding under adverse lighting and motion conditions. The dataset is divided into aligned data and raw data. Each sample in the aligned data contains three modalities: spatially and temporally corresponding RGB images, event representations, and object annotations, and is further divided into training, validation, and test sets. The raw dataset provides unprocessed raw RGB images and raw event streams for custom preprocessing, along with RGB camera and event camera calibration parameters.

Citation

Zhuang, M., Yang, C., Chen, X.& Zhou, J. (2026). Dual-modal Roadside Traffic Dataset (Version 2.0) [Dataset]. Zenodo. https://doi.org/10.5281/zenodo.21070857