Efficient Semantic Splatting for Remote Sensing Multi-view Segmentation
This addresses the problem of computational efficiency for remote sensing applications, though it appears incremental as it builds on existing Gaussian Splatting methods.
The paper tackles efficient multi-view semantic segmentation in remote sensing by proposing a semantic splatting approach based on Gaussian Splatting, which simultaneously renders RGB images and segmentation results from point clouds, achieving significant efficiency gains in optimization and rendering.
In this paper, we propose a novel semantic splatting approach based on Gaussian Splatting to achieve efficient and low-latency. Our method projects the RGB attributes and semantic features of point clouds onto the image plane, simultaneously rendering RGB images and semantic segmentation results. Leveraging the explicit structure of point clouds and a one-time rendering strategy, our approach significantly enhances efficiency during optimization and rendering. Additionally, we employ SAM2 to generate pseudo-labels for boundary regions, which often lack sufficient supervision, and introduce two-level aggregation losses at the 2D feature map and 3D spatial levels to improve the view-consistent and spatial continuity.