AI RESEARCH

RSGMamba: Reliability-Aware Self-Gated State Space Model for Multimodal Semantic Segmentation

arXiv CS.CV

ArXi:2604.12319v1 Announce Type: new Multimodal semantic segmentation has emerged as a powerful paradigm for enhancing scene understanding by leveraging complementary information from multiple sensing modalities (e.g., RGB, depth, and thermal). However, existing cross-modal fusion methods often implicitly assume that all modalities are equally reliable, which can lead to feature degradation when auxiliary modalities are noisy, misaligned, or incomplete.