Skip to content

Latest commit

 

History

History
47 lines (38 loc) · 2.59 KB

README.md

File metadata and controls

47 lines (38 loc) · 2.59 KB
Raw

Pedestrian Intention Estimation (PIE)

pie_annotations



This repository contains code and annotations for the Pedestrian Intention Estimation (PIE) dataset: A. Rasouli, I. Kotseruba, T. Kunic, J.K. Tsotsos, PIE: A Large-Scale Dataset and Models for Pedestrian Intention Estimation, ICCV, 2019 and a series of scripts for visualization and scenario evaluation.

Download videos

Videos are grouped into 6 sets corresponding to different routes driven in Toronto, Canada.

The total size of all video clips is approx. 74GB.

Download links YorkU server Google Drive

Content

  • annotations: Contains the annotations and a script for extracting images from raw videos
  • scenarioEval: Contains code for scenario generation and metrics for trajectory and action prediction
  • visualization: Contains a series of scripts for visualizing data samples and illustrating trajectory and action prediction modules
  • model_outputs: Contains models' outputs for behavior prediction to be used for evaluation and visualization
  • utilities: Contains configuration file for evaluation and visualization, dataset interfaces, and other utility functions
  • camera_params: Contains camera parameters for the PIE dataset

Citation

If you use our dataset, please cite:

@InProceedings{Rasouli_2019_ICCV,
author = {Rasouli, Amir and Kotseruba, Iuliia and Kunic, Toni and Tsotsos, John K.},
title = {PIE: A Large-Scale Dataset and Models for Pedestrian Intention Estimation and Trajectory Prediction},
booktitle = {International Conference on Computer Vision (ICCV)},
year = {2019}}

Authors

Please send an email to yulia_k@eecs.yorku.ca or arasouli.ai@gmail.com if there are any problems with downloading or using the data.

License

This project is licensed under the MIT License - see the LICENSE file for details