Learning Aerial Image Segmentation From Online Maps

By Pascal Kaiser, Jan Dirk Wegner, Aurélien Lucchi, Martin Jaggi, Thomas Hofmann, Konrad Schindler. IEEE Transactions on Geoscience and Remote Sensing, 2017

319 citations recorded by the source corpus.

This paper deals with semantic segmentation of high-resolution (aerial) images where a semantic class label is assigned to each pixel via supervised classification as a basis for automatic map generation. Recently, deep convolutional neural networks (CNNs) have shown impressive performance and have quickly become the de-facto standard for semantic segmentation, with the added benefit that task-specific feature design is no longer necessary. However, a major downside of deep learning methods is that they are extremely data hungry, thus aggravating the perennial bottleneck of supervised classification, to obtain enough annotated training data. On the other hand, it has been observed that they are rather robust against noise in the training labels. This opens up the intriguing possibility to avoid annotating huge amounts of training data, and instead train the classifier from existing legacy data or crowd-sourced maps that can exhibit high levels of noise. The question addressed in this paper is: can training with large-scale publicly available labels replace a substantial part of the manual labeling effort and still achieve sufficient performance? Such data will inevitably contain a significant portion of errors, but in return virtually unlimited quantities of it are available in larger parts of the world. We adapt a state-of-the-art CNN architecture for semantic segmentation of buildings and roads in aerial images, and compare its performance when using different training data sets, ranging from manually labeled pixel-accurate ground truth of the same city to automatic training data derived from OpenStreetMap data from distant locations. We report our results that indicate that satisfying performance can be obtained with significantly less manual annotation effort, by exploiting noisy large-scale training data.

Subfields: Computer Vision and Pattern Recognition, Media Technology, Ocean Engineering

View Paper: https://www.semanticscholar.org/paper/e0cd100dbf93befe1d224f5efe63d670595f59a0

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

In Ocean Engineering

Citations in the corpus

Topological Map Extraction From Overhead Images (2018, background): “Building segmentation from overhead data has been a core research interest for decades and discussing all works is beyond the scope of this paper [19, 34, 20].”
Domain2Vec: Domain Embedding for Unsupervised Domain Adaptation (2020, methodology): “Color [37] 10,000 pedestrian 24 LabelMeFacade [38] 395 building 25 UT Zappos50K [39] 10,000 shoes 26 TRANCOS [40] 1,244 traffic 27 FGVC [40] 10,000 aeroplane 28 Mall Dataset [41] 2,000 mall 29 Chars74K [42] 7,705 character 30 DomainNet [8] 10,000 painting 31 Paris Dataset [43] 3,187 street 32 Domain”