LBTCap: A Lightweight Bilateral Transformer for Real-Time Remote Sensing Image Change Captioning

2026-07-03

AI Summary

This work addresses the high computational cost and low inference efficiency of existing remote sensing image change captioning methods, which hinder real-time deployment. To overcome these limitations, the authors propose LBTCap, a lightweight bilateral Transformer framework that employs a structurally symmetric, learnable bilateral attention mechanism to explicitly model bitemporal features while substantially reducing model size. By integrating a truncated backbone network with grouped-query attention, LBTCap further enhances inference efficiency. The resulting model contains only 39.99 million parameters—of which the change-aware encoder accounts for merely 2.78 million—achieving competitive or near state-of-the-art accuracy while significantly accelerating inference, particularly excelling in low-resource scenarios.

Abstract

Remote sensing image change captioning (RSICC) generates natural-language descriptions of semantic changes between paired remote sensing images (RSIs), supporting applications such as urban planning, disaster response, and environmental monitoring. Although recent methods achieve strong captioning accuracy, most overlook computational efficiency and inference speed, which are essential for real-time deployment in practice. To this end, we propose LBTCap, a lightweight RSICC framework built on a bilateral Transformer that jointly models pre- and post-change features for efficient processing of paired RSIs. Specifically, we introduce a bilateral attention mechanism for paired inputs: the two temporal images are projected into separate queries and keys by the same query and key matrices shared across the two images, the value is formed from their concatenation, and the two resulting attention maps are combined by a learnable, structurally bilateral weighting instead of a fixed subtraction. This design keeps both temporal branches explicit while remaining compact, and, together with a truncated backbone and grouped-query attention, LBTCap uses only 39.99M parameters, of which the change-aware encoder accounts for just 2.78M. Extensive experiments on two public RSICC datasets show that LBTCap matches or closely approaches the accuracy of state-of-the-art methods while using far fewer parameters and running at markedly higher inference speed, with the benefit of the bilateral formulation most pronounced in the low-resource setting, demonstrating a favorable accuracy-efficiency trade-off for practical RSICC.

Problem

• Remote sensing image change captioning
• Computational efficiency
• Inference speed
• Real-time deployment
• Lightweight model

Innovation

• bilateral attention
• lightweight Transformer
• remote sensing image change captioning
• grouped-query attention
• real-time inference