# Bi-Temporal Semantic Reasoning for the Semantic Change Detection in HR Remote Sensing Images

arXiv: https://arxiv.org/abs/2108.06103 · HTML: https://mlsanity.com/paper/2108.06103/

## Key facts

- Cited by 43 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 66 references that are in the map.
- First posted on arXiv 13 Aug 2021.

## Builds on (most cited first, 17 of 17)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1802.02611/) (2018)
- [Non-local Neural Networks](https://mlsanity.com/paper/1711.07971/) (2017)
- [Fully Convolutional Siamese Networks for Change Detection](https://mlsanity.com/paper/1810.08462/) (2018)
- [Land-Cover Classification with High-Resolution Remote Sensing Images Using Transferable Deep Models](https://mlsanity.com/paper/1807.05713/) (2018)
- [Dual Attention Networks for Multimodal Reasoning and Matching](https://mlsanity.com/paper/1611.00471/) (2016)
- [Learning Spectral-Spatial-Temporal Features via a Recurrent Convolutional Neural Network for Change Detection in Multispectral Imagery](https://mlsanity.com/paper/1803.02642/) (2018)
- [GETNET: A General End-to-end Two-dimensional CNN Framework for Hyperspectral Image Change Detection](https://mlsanity.com/paper/1905.01662/) (2019)
- [A Relation-Augmented Fully Convolutional Network for Semantic Segmentation in Aerial Scenes](https://mlsanity.com/paper/1904.05730/) (2019)
- [Semantic Change Detection with Asymmetric Siamese Networks](https://mlsanity.com/paper/2010.05687/) (2020)
- [Deep Image Translation with an Affinity-Based Change Prior for Unsupervised Multimodal Change Detection](https://mlsanity.com/paper/2001.04271/) (2020)
- [Adversarial Shape Learning for Building Extraction in VHR Remote Sensing Images](https://mlsanity.com/paper/2102.11262/) (2021)
- [DiResNet: Direction-aware Residual Network for Road Extraction in VHR Remote Sensing Images](https://mlsanity.com/paper/2005.07232/) (2020)
- [Semantic Change Detection with Hypermaps](https://mlsanity.com/paper/1604.07513/) (2016)

## Built on by (most cited first, 30 of 43)

- [ChangeMamba: Remote Sensing Change Detection With Spatiotemporal State Space Model](https://mlsanity.com/paper/2404.03425/) (2024)
- [Adapting Segment Anything Model for Change Detection in HR Remote Sensing Images](https://mlsanity.com/paper/2309.01429/) (2023)
- [A New Learning Paradigm for Foundation Model-based Remote Sensing Change Detection](https://mlsanity.com/paper/2312.01163/) (2023)
- [Joint Spatio-Temporal Modeling for the Semantic Change Detection in Remote Sensing Images](https://mlsanity.com/paper/2212.05245/) (2022)
- [Change Detection Methods for Remote Sensing in the Last Decade: A Comprehensive Review](https://mlsanity.com/paper/2305.05813/) (2023)
- [Changen2: Multi-Temporal Remote Sensing Generative Change Foundation Model](https://mlsanity.com/paper/2406.17998/) (2024)
- [Change Detection Meets Visual Question Answering](https://mlsanity.com/paper/2112.06343/) (2021)
- [Semantic-CD: Remote Sensing Image Semantic Change Detection towards Open-vocabulary Setting](https://mlsanity.com/paper/2501.06808/) (2025)
- [FUSU: A Multi-temporal-source Land Use Change Segmentation Dataset for Fine-grained Urban Semantic Understanding](https://mlsanity.com/paper/2405.19055/) (2024)
- [ObjFormer: Learning Land-Cover Changes From Paired OSM Data and Optical High-Resolution Imagery via Object-Guided Transformer](https://mlsanity.com/paper/2310.02674/) (2023)
- [UniChange: Unifying Change Detection with Multimodal Large Language Model](https://mlsanity.com/paper/2511.02607/) (2025)
- [RemoteAgent: Bridging Vague Human Intents and Earth Observation with RL-based Agentic MLLMs](https://mlsanity.com/paper/2604.07765/) (2026)
- [Satellite Image Time Series Semantic Change Detection: Novel Architecture and Analysis of Domain Shift](https://mlsanity.com/paper/2407.07616/) (2024)
- [FoBa: A Foreground-Background co-Guided Method and New Benchmark for Remote Sensing Semantic Change Detection](https://mlsanity.com/paper/2509.15788/) (2025)
- [TaCo: Capturing Spatio-Temporal Semantic Consistency in Remote Sensing Change Detection](https://mlsanity.com/paper/2511.20306/) (2025)
- [Referring Change Detection in Remote Sensing Imagery](https://mlsanity.com/paper/2512.11719/) (2025)
- [A Late-Stage Bitemporal Feature Fusion Network for Semantic Change Detection](https://mlsanity.com/paper/2406.10678/) (2024)
- [OmniOVCD: Streamlining Open-Vocabulary Change Detection with SAM 3](https://mlsanity.com/paper/2601.13895/) (2026)
- [Unsupervised CD in satellite image time series by contrastive learning and feature tracking](https://mlsanity.com/paper/2304.11375/) (2023)
- [Foundation Model-Driven Semantic Change Detection in Remote Sensing Imagery](https://mlsanity.com/paper/2602.13780/) (2026)
- [Irregular Change Detection in Sparse Bi-Temporal Point Clouds using Learned Place Recognition Descriptors and Point-to-Voxel Comparison](https://mlsanity.com/paper/2306.15416/) (2023)
- [Mamba-FCS: Joint Spatio- Frequency Feature Fusion, Change-Guided Attention, and SeK Loss for Enhanced Semantic Change Detection in Remote Sensing](https://mlsanity.com/paper/2508.08232/) (2025)
- [SoftMatch Distance: A Novel Distance for Weakly-Supervised Trend Change Detection in Bi-Temporal Images](https://mlsanity.com/paper/2303.04737/) (2023)
- [Hard Region Aware Network for Remote Sensing Change Detection](https://mlsanity.com/paper/2305.19513/) (2023)
- [MapChange: Enhancing Semantic Change Detection with Temporal-Invariant Historical Maps Based on Deep Triplet Network](https://mlsanity.com/paper/2401.11489/) (2024)
- [Denoising of Geodetic Time Series Using Spatiotemporal Graph Neural Networks: Application to Slow Slip Event Extraction](https://mlsanity.com/paper/2405.03320/) (2024)
- [Integrated Dynamic Phenological Feature for Remote Sensing Image Land Cover Change Detection](https://mlsanity.com/paper/2408.04144/) (2024)
- [Detect Changes like Humans: Incorporating Semantic Priors for Improved Change Detection](https://mlsanity.com/paper/2412.16918/) (2024)
- [TASAM: Terrain-and-Aware Segment Anything Model for Temporal-Scale Remote Sensing Segmentation](https://mlsanity.com/paper/2509.15795/) (2025)
- [ChessMamba: Structure-Aware Interleaving of State Spaces for Change Detection in Remote Sensing Images](https://mlsanity.com/paper/2511.19882/) (2025)

## Abstract (arXiv)

Semantic change detection (SCD) extends the multi-class change detection (MCD) task to provide not only the change locations but also the detailed land-cover/land-use (LCLU) categories before and after the observation intervals. This fine-grained semantic change information is very useful in many applications. Recent studies indicate that the SCD can be modeled through a triple-branch Convolutional Neural Network (CNN), which contains two temporal branches and a change branch. However, in this architecture, the communications between the temporal branches and the change branch are insufficient. To overcome the limitations in existing methods, we propose a novel CNN architecture for the SCD, where the semantic temporal features are merged in a deep CD unit. Furthermore, we elaborate on this architecture to reason the bi-temporal semantic correlations. The resulting Bi-temporal Semantic Reasoning Network (Bi-SRNet) contains two types of semantic reasoning blocks to reason both single-temporal and cross-temporal semantic correlations, as well as a novel loss function to improve the semantic consistency of change detection results. Experimental results on a benchmark dataset show that the proposed architecture obtains significant accuracy improvements over the existing approaches, while the added designs in the Bi-SRNet further improves the segmentation of both semantic categories and the changed areas. The codes in this paper are accessible at: github.com/ggsDing/Bi-SRNet.