Semantic Segmentation of Unmanned Aerial Vehicle Remote Sensing Images using SegFormer

TL;DR

This paper evaluates the SegFormer framework for semantic segmentation of UAV remote sensing images, demonstrating its high accuracy and efficiency across different model variants on the UAVid dataset.

Contribution

It provides a comprehensive assessment of SegFormer variants for UAV image segmentation, detailing architecture and training tailored to this application.

Findings

SegFormer models achieve high accuracy in UAV image segmentation.

Real-time and high-performance variants balance speed and precision.

Experimental results outperform existing methods on UAVid dataset.

Abstract

The escalating use of Unmanned Aerial Vehicles (UAVs) as remote sensing platforms has garnered considerable attention, proving invaluable for ground object recognition. While satellite remote sensing images face limitations in resolution and weather susceptibility, UAV remote sensing, employing low-speed unmanned aircraft, offers enhanced object resolution and agility. The advent of advanced machine learning techniques has propelled significant strides in image analysis, particularly in semantic segmentation for UAV remote sensing images. This paper evaluates the effectiveness and efficiency of SegFormer, a semantic segmentation framework, for the semantic segmentation of UAV images. SegFormer variants, ranging from real-time (B0) to high-performance (B5) models, are assessed using the UAVid dataset tailored for semantic segmentation tasks. The research details the architecture and training procedures specific to SegFormer in the context of UAV semantic segmentation. Experimental results showcase the model's performance on benchmark dataset, highlighting its ability to accurately delineate objects and land cover features in diverse UAV scenarios, leading to both high efficiency and performance.