TAMMs: Temporal-Aware Multimodal Model for Satellite Image Change Understanding and Forecasting
This addresses the challenge of disjointed and limited temporal analysis in satellite imagery for applications like environmental monitoring, though it is incremental as it builds on existing MLLM-diffusion architectures.
The paper tackled the problem of modeling long-range temporal dynamics in satellite image analysis by introducing TAMMs, a unified framework for Temporal Change Description and Future Satellite Image Forecasting, which significantly outperformed state-of-the-art baselines on both tasks.
Temporal Change Description (TCD) and Future Satellite Image Forecasting (FSIF) are critical, yet historically disjointed tasks in Satellite Image Time Series (SITS) analysis. Both are fundamentally limited by the common challenge of modeling long-range temporal dynamics. To explore how to improve the performance of methods on both tasks simultaneously by enhancing long-range temporal understanding capabilities, we introduce TAMMs, the first unified framework designed to jointly perform TCD and FSIF within a single MLLM-diffusion architecture. TAMMs introduces two key innovations: Temporal Adaptation Modules (TAM) enhance frozen MLLM's ability to comprehend long-range dynamics, and Semantic-Fused Control Injection (SFCI) mechanism translates this change understanding into fine-grained generative control. This synergistic design makes the understanding from the TCD task to directly inform and improve the consistency of the FSIF task. Extensive experiments demonstrate TAMMs significantly outperforms state-of-the-art specialist baselines on both tasks.