Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces FU-Mamba, a novel framework for oralscan image segmentation that addresses the limitations of existing visual state space models by incorporating dynamic scanning and frequency domain enhancement. The framework utilizes a Dynamic Mamba Block (DMB) to learn adaptive sampling offsets and perform flexible bilinear interpolation, thereby maintaining spatial coherence during segmentation. Experimental results demonstrate a significant improvement in segmentation accuracy, achieving a 1.1% increase in the mean intersection over union (mIoU) metric on a dental segmentation dataset, highlighting its effectiveness under challenging imaging conditions.
FU-Mamba boosts segmentation accuracy by 1.1% mIoU through innovative dynamic scanning and frequency enhancement techniques, tackling common pitfalls in dental imaging.
Oralscan image segmentation is essential for computer-aided diagnosis and treatment planning in digital dentistry. However, existing visual state space models (SSMs) often rely on manually designed scanning orders to flatten image patches into sequences, which disrupts the semantic spatial continuity and hinders coherent feature extraction from key foreground regions. Moreover, elements such as inconsistent lighting, reflective surfaces, and noise during data acquisition disrupt the frequency distribution by diminishing high-frequency details while enhancing low-frequency components, consequently hindering the accurate localization of boundaries. In response to these challenges, we introduce FU-Mamba, an innovative framework that incorporates dynamic scanning and frequency domain enhancement within the SSM architecture. Specifically, the Dynamic Mamba Block (DMB) adaptively learns sampling offsets via a trainable offset prediction network and performs flexible bilinear interpolation, enabling content-aware scanning that preserves spatial coherence. Furthermore, a frequency domain enhancement block balances spectral components through wavelet-guided decomposition and spectrum pooling, improving robustness under adverse imaging conditions. Experimental findings indicate that FU-Mamba attains a notable enhancement in segmentation accuracy, evidenced by a 1.1% increase in the mean intersection over union (mIoU) metric when evaluated on the dental segmentation dataset. Project page: https://byte2bite.github.io/FU-Mamba/