RaDe-GS: Rasterizing Depth in Gaussian Splatting

arXiv:2406.01467v1 Announce Type: cross
Abstract: Gaussian Splatting (GS) has proven to be highly effective in novel view synthesis, achieving high-quality and real-time rendering. However, its potential for reconstructing detailed 3D shapes has not been fully explored. Existing methods often suffer from limited shape accuracy due to the discrete and unstructured nature of Gaussian splats, which complicates the shape extraction. While recent techniques like 2D GS have attempted to improve shape reconstruction, they often reformulate the Gaussian primitives in ways that …

abstract accuracy arxiv cs.cv cs.gr however nature novel potential quality real-time rendering shape synthesis type unstructured view