Automatic Pixelwise Object Labeling for Aerial Imagery Using Stacked U-Nets

TL;DR

This paper introduces a stacked U-Net pipeline for automatic pixelwise object labeling in aerial imagery, demonstrating improved accuracy and efficiency across diverse datasets and scalable processing methods.

Contribution

The paper presents a novel end-to-end stacked U-Net architecture that outperforms existing methods and enables efficient processing through sub-sampling without significant quality loss.

Findings

Outperforms state-of-the-art on two aerial datasets

Enables faster processing with minimal quality degradation

Generalizable approach applicable to various image types

Abstract

Automation of objects labeling in aerial imagery is a computer vision task with numerous practical applications. Fields like energy exploration require an automated method to process a continuous stream of imagery on a daily basis. In this paper we propose a pipeline to tackle this problem using a stack of convolutional neural networks (U-Net architecture) arranged end-to-end. Each network works as post-processor to the previous one. Our model outperforms current state-of-the-art on two different datasets: Inria Aerial Image Labeling dataset and Massachusetts Buildings dataset each with different characteristics such as spatial resolution, object shapes and scales. Moreover, we experimentally validate computation time savings by processing sub-sampled images and later upsampling pixelwise labeling. These savings come at a negligible degradation in segmentation quality. Though the conducted experiments in this paper cover only aerial imagery, the technique presented is general and can handle other types of images.