Unsupervised Domain Adaptation from Synthetic to Real Images for Anchorless Object Detection

Tobias Scheck, Ana Grassi, Gangolf Hirtz

Abstract

Synthetic images are one of the most promising solutions to avoid high costs associated with generating annotated datasets to train supervised convolutional neural networks (CNN). However, to allow networks to generalize knowledge from synthetic to real images, domain adaptation methods are necessary. This paper implements unsupervised domain adaptation (UDA) methods on an anchorless object detector. Given their good performance, anchorless detectors are increasingly attracting attention in the field of object detection. While their results are comparable to the well-established anchor-based methods, anchorless detectors are considerably faster. In our work, we use CenterNet, one of the most recent anchorless architectures, for a domain adaptation problem involving synthetic images. Taking advantage of the architecture of anchorless detectors, we propose to adjust two UDA methods, viz., entropy minimization and maximum squares loss, originally developed for segmentation, to object detection. Our results show that the proposed UDA methods can increase the mAP from 61% to 69% with respect to direct transfer on the considered anchorless detector. The code is available: https://github.com/scheckmedia/centernet-uda.

Download


Paper Citation


in Harvard Style

Scheck T., Grassi A. and Hirtz G. (2021). Unsupervised Domain Adaptation from Synthetic to Real Images for Anchorless Object Detection.In Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP, ISBN 978-989-758-488-6, pages 319-327. DOI: 10.5220/0010202503190327


in Bibtex Style

@conference{visapp21,
author={Tobias Scheck and Ana Grassi and Gangolf Hirtz},
title={Unsupervised Domain Adaptation from Synthetic to Real Images for Anchorless Object Detection},
booktitle={Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP,},
year={2021},
pages={319-327},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010202503190327},
isbn={978-989-758-488-6},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP,
TI - Unsupervised Domain Adaptation from Synthetic to Real Images for Anchorless Object Detection
SN - 978-989-758-488-6
AU - Scheck T.
AU - Grassi A.
AU - Hirtz G.
PY - 2021
SP - 319
EP - 327
DO - 10.5220/0010202503190327