MAPLE: Multi-Path Adaptive Propagation with Level-Aware Embeddings for Hierarchical Multi-Label Image Classification
This addresses the challenge of modeling structured label dependencies in remote sensing for Earth observation applications, representing an incremental advance with specific gains.
The paper tackles the problem of hierarchical multi-label classification in remote sensing, where existing methods underuse hierarchical information in multi-path settings, and proposes MAPLE, which improves performance by up to +42% in few-shot regimes with only a 2.6% parameter overhead.
Hierarchical multi-label classification (HMLC) is essential for modeling structured label dependencies in remote sensing. Yet existing approaches struggle in multi-path settings, where images may activate multiple taxonomic branches, leading to underuse of hierarchical information. We propose MAPLE (Multi-Path Adaptive Propagation with Level-Aware Embeddings), a framework that integrates (i) hierarchical semantic initialization from graph-aware textual descriptions, (ii) graph-based structure encoding via graph convolutional networks (GCNs), and (iii) adaptive multi-modal fusion that dynamically balances semantic priors and visual evidence. An adaptive level-aware objective automatically selects appropriate losses per hierarchy level. Evaluations on CORINE-aligned remote sensing datasets (AID, DFC-15, and MLRSNet) show consistent improvements of up to +42% in few-shot regimes while adding only 2.6% parameter overhead, demonstrating that MAPLE effectively and efficiently models hierarchical semantics for Earth observation (EO).