Evaluating Deep Learning Architectures and Transferability for Building Segmentation from UAV Imagery in Kathmandu Valley

High-resolution UAV imagery lets municipalities map building footprints where no cadastral record exists, and Nepal, among the fastest urbanizing countries, has little such data. Public benchmarks cover Europe, North America, and East Asia, and UAV datasets are mostly private or limited to Chinese cities, so South Asian and Himalayan settlements are uncovered. No study has compared convolutional, transformer, and hybrid architectures across a six-fold ground sampling distance gap, or tested transfer between two settlements of one valley. We annotated two new UAV datasets over Khokana (0.15 m/px) and Lamatar (0.026 m/px), peri-urban settlements in Lalitpur, Nepal, and resampled Lamatar to a common 0.15 m/px grid. We trained a U-Net with ResNet encoders, SegFormer-B2, and TransResUNet from scratch, with full fine-tuning, and with a frozen encoder, then transferred the best model per site in both directions. SegFormer gave the highest within-site IoU (85.9% Lamatar, 80.7% Khokana) and transferred into Lamatar at 82.5% with no adaptation. Transfer into Khokana began at 65.7% and needed full fine-tuning, which lifted all three architectures past their Khokana ceilings. Agencies without cadastral data can annotate one area in full and adapt to a second from a small sample, and the transfer direction determines the strategy.

Authors

Institutions

Publication Details

Published
2026-11-17
DOI
https://doi.org/10.22541/authorea.15012367/v1
Primary Topic
Automated Road and Building Extraction
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Evaluating Deep Learning Architectures and Transferability for Building Segmentation from UAV Imagery in Kathmandu Valley

Bishnu Khadka, Sunil Thapa, Niranjan Khakurel, Anupam Shrestha et al.
5 links
Automated Road and Building Extraction
preprint

Evaluating Deep Learning Architectures and Transferability for Building Segmentation from UAV Imagery in Kathmandu Valley

Bishnu Khadka, Sunil Thapa, Niranjan Khakurel, Anupam Shrestha, Anup Shrestha, Bipun Pati, Rohan Shrestha, Rusha Manandhar
preprint en

Abstract

High-resolution UAV imagery lets municipalities map building footprints where no cadastral record exists, and Nepal, among the fastest urbanizing countries, has little such data. Public benchmarks cover Europe, North America, and East Asia, and UAV datasets are mostly private or limited to Chinese cities, so South Asian and Himalayan settlements are uncovered. No study has compared convolutional, transformer, and hybrid architectures across a six-fold ground sampling distance gap, or tested transfer between two settlements of one valley. We annotated two new UAV datasets over Khokana (0.15 m/px) and Lamatar (0.026 m/px), peri-urban settlements in Lalitpur, Nepal, and resampled Lamatar to a common 0.15 m/px grid. We trained a U-Net with ResNet encoders, SegFormer-B2, and TransResUNet from scratch, with full fine-tuning, and with a frozen encoder, then transferred the best model per site in both directions. SegFormer gave the highest within-site IoU (85.9% Lamatar, 80.7% Khokana) and transferred into Lamatar at 82.5% with no adaptation. Transfer into Khokana began at 65.7% and needed full fine-tuning, which lifted all three architectures past their Khokana ceilings. Agencies without cadastral data can annotate one area in full and adapt to a second from a small sample, and the transfer direction determines the strategy.

Tribhuvan University (NP), Kathmandu College of Management (NP), Nepal College of Information Technology (NP), Institute of Engineering Thapathali Campus (NP), Nepal Academy of Science and Technology (NP)
Sustainable cities and communities, Industry, innovation and infrastructure
Automated Road and Building Extraction
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon