English

IntFormer: Predicting pedestrian intention with the aid of the Transformer architecture

Computer Vision and Pattern Recognition 2021-05-19 v1 Artificial Intelligence

Abstract

Understanding pedestrian crossing behavior is an essential goal in intelligent vehicle development, leading to an improvement in their security and traffic flow. In this paper, we developed a method called IntFormer. It is based on transformer architecture and a novel convolutional video classification model called RubiksNet. Following the evaluation procedure in a recent benchmark, we show that our model reaches state-of-the-art results with good performance (40\approx 40 seq. per second) and size (8×8\times smaller than the best performing model), making it suitable for real-time usage. We also explore each of the input features, finding that ego-vehicle speed is the most important variable, possibly due to the similarity in crossing cases in PIE dataset.

Keywords

Cite

@article{arxiv.2105.08647,
  title  = {IntFormer: Predicting pedestrian intention with the aid of the Transformer architecture},
  author = {J. Lorenzo and I. Parra and M. A. Sotelo},
  journal= {arXiv preprint arXiv:2105.08647},
  year   = {2021}
}

Comments

5 pages, 2 figures