Data underlying the MSc Thesis: Toward Occlusion Capable Human Trajectory Prediction

Here are the dataset and model files related to the MSc thesis: Toward Occlusion Capable Human Trajectory Prediction.This thesis focuses on handling occlusions and partially missing positional information of agents when predicting their trajectories.<br>Dataset files comprise three distinct versions of the trajectory dataset used throughout the thesis project:train, val and test splits for the trajectory dataset under <em>fully observed</em> conditionstrain, val and test splits for the trajectory dataset under <em>occluded</em> conditionstest split for the trajectory dataset under occluded conditions, with <em>imputation </em>of missing positions by means of interpolation and constant velocity extrapolation.Occluded trajectories were generated by applying a simulator of occlusion events onto a publicly available trajectory dataset: the Stanford Drone Dataset. Our dataset Files are therefore derived from the Stanford Drone Dataset.<br>Model files contain checkpoints with weights that can be used to initialize prediction models from our implementation. These checkpoints are accompanied by some metadata, with information about the evolution of train and validation losses throughout the training process of individual model instances. Models are trained in two separate phases (<em>I</em>, and <em>II</em>): each model file contains all relevant model data for both phases of one model instance.<br>The source code related to these files is hosted on this GitHub page. The repository's README contains a comprehensive set of instructions on how to use the files in order to reproduce the results we obtained in our research (please refer to the section titled "Downloading Models and Legacy Datasets").