"Rainy Pedestrian Trajectory Dataset (RPTD)"

"This dataset provides rain-degraded pedestrian trajectory data for evaluating the robustness of trajectory prediction models under adverse observation conditions. The dataset is constructed by introducing simulated rainy weather degradation into widely used pedestrian trajectory benchmarks while preserving the original temporal consistency and motion structure of trajectories.The dataset is designed to support research on robust trajectory forecasting, degraded observation modeling, and geometry-aware motion prediction. By incorporating realistic observation corruption caused by rain, reduced visibility, and sensor noise, the dataset enables systematic evaluation of trajectory prediction algorithms under challenging environmental conditions.This dataset can be used for benchmarking models in pedestrian trajectory prediction, especially for studying robustness under noisy or incomplete observations. It is suitable for training, validation, and testing of machine learning and deep learning models for motion prediction tasks."