S2Looking

Dataset Information
Modalities
Images
Introduced
2021
License
Unknown
Homepage

Overview

S2Looking is a building change detection dataset that contains large-scale side-looking satellite images captured at varying off-nadir angles. The S2Looking dataset consists of 5,000 registered bitemporal image pairs (size of 1024*1024, 0.5 ~ 0.8 m/pixel) of rural areas throughout the world and more than 65,920 annotated change instances. We provide two label maps to separately indicate the newly built and demolished building regions for each sample in the dataset. We establish a benchmark task based on this dataset, i.e., identifying the pixel-level building changes in the bi-temporal images.

Variants: S2Looking

Associated Benchmarks

This dataset is used in 1 benchmark:

Recent Benchmark Submissions

Task Model Paper Date
Change Detection SChanger-base SChanger: Change Detection from a … 2025-03-26
Change Detection SChanger-small SChanger: Change Detection from a … 2025-03-26
Change Detection C2FNet C2F-SemiCD: A Coarse-to-Fine Semi-Supervised Change … 2024-04-22
Change Detection HANet HANet: A Hierarchical Attention Network … 2024-04-14
Change Detection CGNet Change Guiding Network: Incorporating Change … 2024-04-14
Change Detection LSKNet-S LSKNet: A Foundation Lightweight Backbone … 2024-03-18
Change Detection BD-MSA BD-MSA: Body decouple VHR Remote … 2024-01-09
Change Detection ChangeStar(1x96, MiT-b1, Changen-90k) Scalable Multi-Temporal Remote Sensing Change … 2023-09-29
Change Detection ChangeStar(1x96, ResNet-18, Changen-90k) Scalable Multi-Temporal Remote Sensing Change … 2023-09-29
Change Detection HCGMNet HCGMNET: A Hierarchical Change Guiding … 2023-02-21
Change Detection Changer (Ex, ResNet-18) Changer: Feature Interaction is What … 2022-09-17

Research Papers

Recent papers with results on this dataset: