NeuroMamba: A Universal Spatiotemporal Module for Robust Perception in Degraded Sensory Streams

In open-world intelligent systems, processing continuous sensory streams disrupted by heterogeneous degradation sources presents a fundamental challenge: reconciling the inherent tension between observational completeness and reconstruction fidelity. Methods that prioritize completeness by bridging long-term occlusions often introduce spurious artifacts, while approaches focused on aggressive noise suppression inevitably disrupt temporal continuity and erase valid structures. To address this challenge, we propose NeuroMamba, a universal plug-and-play module that enhances spatiotemporal consistency in degraded streams. NeuroMamba tackles the dual objectives through two synergistic components. First, we propose a Regional Hybrid Spatiotemporal Rectification (HSR) module, which leverages the linear complexity O(L) of Mamba-based inertial modeling to recover long-range temporal dependencies and infer missing modalities under partial observability. Second, we introduce a Spiking Confidence Gate (SCG) that enforces reconstruction fidelity via physics-guided supervision. Acting as a hard neuromorphic filter governed by integrate-and-fire (LIF) dynamics, SCG distinguishes valid geometric features from sensor noise based on accumulated temporal evidence. Extensive experiments on the nuScenes robustness benchmark demonstrate that NeuroMamba effectively reconciles the completeness-fidelity trade-off, achieving state-of-the-art performance in restoring high-fidelity spatiotemporal features from severely incomplete and degraded observations.