EDGS: Eliminating Densification for Efficient Convergence of 3DGS

EDGS: Eliminating Densification for Efficient Convergence of 3DGS

GitHub – CompVis/EDGS: [CVPR 2026] A PyTorch implementation of the paper “EDGS: Eliminating Densification for Efficient Convergence of 3DGS”

“3DGS initializes with a sparse set of Gaussians and progressively adds more in under-reconstructed regions. In contrast, EDGS starts with a dense initialization from triangulated 2D correspondences across training image pairs, requiring only minimal refinement. This leads to faster convergence and higher rendering quality. Our method reaches the original 3DGS LPIPS score in just 25% of the training time and uses only 60% of the splats. Renderings become nearly indistinguishable from ground truth after only 3,000 steps — without any densification…”

Source: github.com/CompVis/EDGS