Contrastive Pretraining for Railway Detection: Unveiling Historical Maps with Transformers
METADATA ONLY
Loading...
Author / Producer
Date
2023-11
Publication Type
Conference Paper
ETH Bibliography
yes
Citations
Altmetric
METADATA ONLY
Data
Rights / License
Abstract
Detecting railways from historical maps is challenging due to their infrequent representation in a map sheet and their visual similarity with roads. Basically, both railways and roads are symbolised as two parallel black lines, with slight differences only in line thickness. Recent advancements in transformer models for computer vision tasks have sparked interest in utilizing them for processing historical maps. However, the success of transformers heavily relies on large-scale labelled datasets, predominantly available for ground imagery rather than historical maps. To overcome these challenges, we exploit the unique spatial characteristics of historical map data, where the same location can be depicted over different time spans across different map series. For example, each location in Switzerland is depicted in both the Siegfried map and the Old National map, each exhibiting distinct symbols and drawing styles. In this work, we address the scarcity of labelled data by generating positive pairs of the same scene from different map series and employ self-supervised contrastive learning to pre-train a dedicated transformer encoder optimized for map data. Subsequently, we finetune the entire transformer network for the downstream railway detection task. Experimental results demonstrate that our method achieves comparable performance to fully supervised approaches, while significantly reducing the amount of required labelled dataset to a mere 2.5% after contrastive pretraining.
Permanent link
Publication status
published
External links
Book title
GeoAI '23: Proceedings of the 6th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery
Journal / series
Volume
Pages / Article No.
30 - 33
Publisher
Association for Computing Machinery
Event
6th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery (GeoAI 2023)
Edition / version
Methods
Software
Geographic location
Date collected
Date created
Subject
Railway detection; Map processing; Contrastive learning; Transformer; Neural networks
Organisational unit
03466 - Hurni, Lorenz / Hurni, Lorenz
Notes
Funding
192018 - EMPHASES: Assessing EMergent PHenomenA in complex Social-Ecological Systems with time series of settlement and habitat networks (SNF)