English

An RNN-based IMM Filter Surrogate

Computer Vision and Pattern Recognition 2019-04-29 v2

Abstract

The problem of varying dynamics of tracked objects, such as pedestrians, is traditionally tackled with approaches like the Interacting Multiple Model (IMM) filter using a Bayesian formulation. By following the current trend towards using deep neural networks, in this paper an RNN-based IMM filter surrogate is presented. Similar to an IMM filter solution, the presented RNN-based model assigns a probability value to a performed dynamic and, based on them, puts out a multi-modal distribution over future pedestrian trajectories. The evaluation is done on synthetic data, reflecting prototypical pedestrian maneuvers.

Cite

@article{arxiv.1902.01739,
  title  = {An RNN-based IMM Filter Surrogate},
  author = {Stefan Becker and Ronny Hug and Wolfgang Hübner and Michael Arens},
  journal= {arXiv preprint arXiv:1902.01739},
  year   = {2019}
}

Comments

Accepted at Scandinavian Conference on Image Analysis (SCIA) 2019

R2 v1 2026-06-23T07:32:35.797Z