Introducing A Framework for Single-Human Tracking Using Event-Based Cameras - Equipe System on Chip
Conference Papers Year : 2023

Introducing A Framework for Single-Human Tracking Using Event-Based Cameras

Abstract

Event cameras generate data based on the amount of motion present in the captured scene, making them attractive sensors for solving ob- ject tracking tasks. In this paper, we present a framework for tracking humans using a single event camera which consists of three compo- nents. First, we train a Graph Neural Network (GNN) to recognize a person within the stream of events. Batches of events are represented as spatio-temporal graphs in order to preserve the sparse nature of events and retain their high temporal resolution. Subsequently, the person is localized in a weakly-supervised manner by adopting the well established method of Class Activation Maps (CAM) for our graph-based classification model. Our approach does not require the ground truth position of humans during training. Finally, a Kalman filter is deployed for tracking, which uses the predicted bounding box surrounding the human as measurement. We demonstrate that our approach achieves robust tracking results on test sequences from the Gait3 database, paving the way for further privacy-preserving methods in event-based human tracking. Code, pre-trained models and datasets of our research are publicly available
Fichier principal
Vignette du fichier
publi-7427.pdf (352.49 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04255949 , version 1 (24-10-2023)

Identifiers

Cite

Dominik Eisl, Fabian Herzog, Jean-Luc Dugelay, Ludovic Apvrille, Gerhard Rigoll. Introducing A Framework for Single-Human Tracking Using Event-Based Cameras. 2023 IEEE International Conference on Image Processing (ICIP), Oct 2023, Kuala Lumpur, Malaysia. pp.3269-3273, ⟨10.1109/ICIP49359.2023.10222777⟩. ⟨hal-04255949⟩
254 View
38 Download

Altmetric

Share

More