Domain Adaptive Pedestrian Detection Based on Semantic Concepts

Patrick Feifel, Patrick Feifel, Frank Bonarens, Frank Köster, Frank Köster

2023

Abstract

Pedestrian detection is subject to high complexity with a wide variety of pedestrian appearances and postures as well as environmental conditions. Building a sufficient real-world dataset is labor-intensive and costly. Thus, the application of synthetic data is promising, but deep neural networks show a lack of generalization when trained solely on synthetic data. In our work, we propose a novel method for concept-based domain adaptation for pedestrian detection (ConDA). In addition to the 2D bounding box prediction, an auxiliary body part segmentation exploits discriminative features of semantic concepts of pedestrians. Inspired by approaches to the inherent interpretability of DNNs, ConDA has been shown to strengthen generalization. This is done by enforcing a high intra-class concentration and inter-class separation of extracted body part features in the latent space. We report performance results regarding various training strategies, feature extractions and backbones for ConDA on the real-world CityPersons dataset.

Paper Citation

in Harvard Style

Feifel P., Bonarens F. and Köster F. (2023). Domain Adaptive Pedestrian Detection Based on Semantic Concepts. In Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP; ISBN 978-989-758-634-7, SciTePress, pages 652-659. DOI: 10.5220/0011690900003417

in Bibtex Style

@conference{visapp23,
author={Patrick Feifel and Frank Bonarens and Frank Köster},
title={Domain Adaptive Pedestrian Detection Based on Semantic Concepts},
booktitle={Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP},
year={2023},
pages={652-659},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011690900003417},
isbn={978-989-758-634-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP
TI - Domain Adaptive Pedestrian Detection Based on Semantic Concepts
SN - 978-989-758-634-7
AU - Feifel P.
AU - Bonarens F.
AU - Köster F.
PY - 2023
SP - 652
EP - 659
DO - 10.5220/0011690900003417
PB - SciTePress