Saved in:
Bibliographic Details
Main Authors: Kikuchi, Satoshi, Kato, Masaya, Tasaki, Tsuyoshi
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.26170
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918253704511488
author Kikuchi, Satoshi
Kato, Masaya
Tasaki, Tsuyoshi
author_facet Kikuchi, Satoshi
Kato, Masaya
Tasaki, Tsuyoshi
contents Self-localization on a 3D map by using an inexpensive monocular camera is required to realize autonomous driving. Self-localization based on a camera often uses a convolutional neural network (CNN) that can extract local features that are calculated by nearby pixels. However, when dynamic obstacles, such as people, are present, CNN does not work well. This study proposes a new method combining CNN with Vision Transformer, which excels at extracting global features that show the relationship of patches on whole image. Experimental results showed that, compared to the state-of-the-art method (SOTA), the accuracy improvement rate in a CG dataset with dynamic obstacles is 1.5 times higher than that without dynamic obstacles. Moreover, the self-localization error of our method is 20.1% smaller than that of SOTA on public datasets. Additionally, our robot using our method can localize itself with 7.51cm error on average, which is more accurate than SOTA.
format Preprint
id arxiv_https___arxiv_org_abs_2510_26170
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Self-localization on a 3D map by fusing global and local features from a monocular camera
Kikuchi, Satoshi
Kato, Masaya
Tasaki, Tsuyoshi
Robotics
Computer Vision and Pattern Recognition
Self-localization on a 3D map by using an inexpensive monocular camera is required to realize autonomous driving. Self-localization based on a camera often uses a convolutional neural network (CNN) that can extract local features that are calculated by nearby pixels. However, when dynamic obstacles, such as people, are present, CNN does not work well. This study proposes a new method combining CNN with Vision Transformer, which excels at extracting global features that show the relationship of patches on whole image. Experimental results showed that, compared to the state-of-the-art method (SOTA), the accuracy improvement rate in a CG dataset with dynamic obstacles is 1.5 times higher than that without dynamic obstacles. Moreover, the self-localization error of our method is 20.1% smaller than that of SOTA on public datasets. Additionally, our robot using our method can localize itself with 7.51cm error on average, which is more accurate than SOTA.
title Self-localization on a 3D map by fusing global and local features from a monocular camera
topic Robotics
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.26170