Dynamic and Asymmetric Enhancement for Remote Sensing Image Change Captioning
- 21 November 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-15
Abstract
Remote Sensing Image Change Captioning (RSICC) plays a critical role in automated environmental monitoring by generating natural language descriptions that provide intuitive interpretations of changes between bi-temporal remote sensing images. Despite recent advancements, existing methods suffer from two fundamental limitations: (1) static encoding architectures fail to account for scene complexity and semantic diversity during feature extraction, leading to inflexible representation learning; and (2) symmetric computational structures are inherently unsuitable for modeling the asymmetric temporal dependencies inherent in “before-to-after” image pairs. To address these challenges, we propose a Dynamic Asymmetric Encoder (DAE), which introduces two key innovations. First, we design a difference-guided dynamic convolution module that adaptively adjusts convolutional parameters using input-driven scaling factors and offsets, thereby enabling scene-aware intra-image feature enhancement. Second, we develop a Multi-expert Temporal Interaction (METI) module that establishes an asymmetric computational topology: the “after” image branch actively perceives change information relative to the “before” image through three heterogeneous interaction experts, followed by feature fusion. This design allocates greater computational capacity to the “after” image branch while preserving temporal coherence. Furthermore, we introduce a Multi-level Feature Aggregator (MFA) that enhances the representation of salient changed regions across multiple scales via an iterative reinforcement mechanism. Experimental results validate the effectiveness of the proposed method, demonstrating state-of-the-art performance on two benchmark RSICC datasets. We publicly release our code repository at https://github.com/TTXiann/Dynamic-Asymmetric to facilitate future research.Keywords
Funding Information
- National Key Research and Development Program of China (2022YFF0607001)
- Basic and Applied Basic Research Foundation of Guangdong Province (2023A1515010993, 24202107190000687)
- in part by Guangzhou City Science and Technology Research Projects (2023B01J0011)
- in part by Shaoguan Science and Technology Research Project (230316116276286)
- in part by Foshan Science and Technology Research Project (2220001018608)
- in part by Zhuhai Science and Technology Research Project (2320004002668)
- Zhongshan Science and Technology Research Project (2024A1010)
- National Natural Science Foundation of China (32572189)
This publication has 43 references indexed in Scilit:
- Distractors-Immune Representation Learning with Cross-Modal Contrastive Regularization for Change Captioning2024
- Mining core information by evaluating semantic importance for unpaired image captioningNeural Networks, 2024
- MultiMAE: Multi-modal Multi-task Masked Autoencoders2022
- Scaling Up Vision-Language Pretraining for Image Captioning2022
- A ConvNet for the 2020s2022
- Injecting Semantic Concepts into End-to-End Image Captioning2022
- Dual Global Enhanced Transformer for image captioningNeural Networks, 2022
- Meshed-Memory Transformer for Image Captioning2020
- CIDEr: Consensus-based image description evaluation2015
- PCA‐based land‐use change detection and analysis using multitemporal and multisensor satellite dataInternational Journal of Remote Sensing, 2008