Semantic Segmentation from Remote Sensor Data and the Exploitation of Latent Learning for Classification of Auxiliary Tasks

TL;DR

This paper introduces ICT-Net for semantic segmentation of buildings from remote sensor data, demonstrates its superior performance, explores the relationship between classification and reconstruction accuracy, and investigates latent learning for auxiliary task classification.

Contribution

It presents a new deep neural network architecture for building segmentation, analyzes the link between classification and reconstruction accuracy, and introduces the concept of latent learning in deep networks.

Findings

ICT-Net outperforms state-of-the-art by over 1.5% and 1.8% on Jaccard index.

Latent learning enables networks trained on primary tasks to learn auxiliary tasks.

Achieved average F1 scores of 54.29% for roads and 42.74% for low vegetation.

Abstract

In this paper we address three different aspects of semantic segmentation from remote sensor data using deep neural networks. Firstly, we focus on the semantic segmentation of buildings from remote sensor data and propose ICT-Net. The proposed network has been tested on the INRIA and AIRS benchmark datasets and is shown to outperform all other state of the art by more than 1.5% and 1.8% on the Jaccard index, respectively. Secondly, as the building classification is typically the first step of the reconstruction process, we investigate the relationship of the classification accuracy to the reconstruction accuracy. Finally, we present the simple yet compelling concept of latent learning and the implications it carries within the context of deep learning. We posit that a network trained on a primary task (i.e. building classification) is unintentionally learning about auxiliary tasks (e.g. the classification of road, tree, etc) which are complementary to the primary task. We extensively tested the proposed technique on the ISPRS benchmark dataset which contains multi-label ground truth, and report an average classification accuracy (F1 score) of 54.29% (SD=17.03) for roads, 10.15% (SD=2.54) for cars, 24.11% (SD=5.25) for trees, 42.74% (SD=6.62) for low vegetation, and 18.30% (SD=16.08) for clutter. The source code and supplemental material is publicly available at http://www.theICTlab.org/lp/2019ICT-Net/.