Search papers, labs, and topics across Lattice.
This paper introduces CASA-SDF, a novel framework for neural implicit surface reconstruction that tackles the challenges posed by geometric heterogeneity in indoor scenes. By integrating Hybrid Spatially-Adaptive Uncertainty Annealing (SAUA) and Curvature-Aware Locally Adaptive Density Transformation (CALADT), the method effectively balances the need for regularization in texture-less regions with the preservation of detail in thin structures. Experimental results on benchmark datasets show that CASA-SDF significantly enhances surface completeness and detail recovery while maintaining stability in planar surfaces.
CASA-SDF achieves superior indoor surface reconstruction by intelligently balancing regularization and detail preservation, outperforming existing methods.
Neural implicit representations have emerged as a powerful paradigm for 3D reconstruction. However, high-fidelity indoor surface reconstruction remains a significant challenge, primarily due to the pronounced \emph{geometric heterogeneity} of indoor scenes. Large texture-less planar regions typically require stronger regularization to suppress high-frequency artifacts, while thin structures demand sharper, more adaptive representations to mitigate the spectral bias of multi-layer perceptrons (MLPs) and prevent over-smoothing. Existing approaches often rely on spatially indiscriminate prior supervision and a scene-global SDF-to-density transformation, which constrains their ability to balance planar smoothness and detail preservation. In this paper, we propose CASA-SDF (Curriculum-Aware Spatial Adaptation for SDF), a unified framework that addresses this challenge via complementary adaptations of supervision and representation capacity. Specifically, Hybrid Spatially-Adaptive Uncertainty Annealing (SAUA) fuses semantic and photometric uncertainties to construct a pixel-wise curriculum for monocular prior supervision. This strategy maintains regularization in reliable regions while attenuating unreliable supervision early in training to enable data-driven photometric refinement. Meanwhile, Curvature-Aware Locally Adaptive Density Transformation (CALADT) progressively modulates the sharpness of the SDF-to-density mapping via a curvature proxy to enhance the representation of thin structures. Extensive experiments on benchmark indoor datasets demonstrate that CASA-SDF improves surface completeness and detail recovery on high-frequency structures, without compromising the stability of planar surfaces.
