Pedestrians’ red-light crossing can present a threat to traffic safety. Among all the existing work related to pedestrian’s red-light crossing, there are few studies using trajectory data in time sequence. This paper uses pose estimation (keypoint detection) to generate pedestrians’ variables from CCTV videos. Four machine learning models are used to predict pedestrians’ crossing intention at intersections’ red-light. The best model achieves an accuracy of 0.920 and AUC value of 0.849, with data from three intersections. Different prediction horizons (up to 4 sec) are used. With longer prediction horizons, the sample size gets smaller, which partially leads to worse model performance. However, the performance with prediction horizon up to 2 sec is still good (AUC value as 0.841). It is found that keypoint variables such as the angles between ankle and knee (left side) and elbow and shoulder (right side) are important. This model can be further implemented in the Infrastructure-to-Vehicle (I2V) applications and thus prevent accidents due to pedestrians’ red-light crossing by issuing warnings to drivers.

Inclement weather conditions, particularly heavy rain and the consequent wet road surface, have an unfavorable effect on driving conditions, traffic infrastructure, and operational plans. To mitigate the potentially detrimental ramifications of turbulent weather, it must be continuously monitored in real time and with high geospatial granularity. Traditionally, road weather conditions are monitored using weather forecasts or Roadside Weather Information Systems (RWIS). However, these methods are either ill-equipped or too expensive to provide the required fine-grained observations. Alternatively, roadside traffic CCTV cameras are ubiquitously deployed on US freeways and can serve as inexpensive sensors to surveil weather. In this paper, a novel vision-based methodology is proposed to detect rain and road surface conditions from roadside traffic cameras. Vision Transformers were utilized for image-based classification. They demonstrated superior results compared to convolution-based approaches. Furthermore, the geographical distribution of roadside cameras was leveraged to add spatial context awareness to the detection model. A Spatial Self-Attention network was proposed to model the relationship between the detection results of adjacent images as a sequence-to-sequence detection task. The results indicate that the addition of the sequential detection module improved the accuracy of the stand-alone Vision Transformer as measured by the F1-score. The boost in performance enhanced the F1-scores of the stand-alone Vision Transformer by 5.61% and 5.97% for the rain and road surface condition detection tasks, respectively, raising the total F1-score to 96.71% and 98.07%.