A shared MLP predicts Gaussian visibility from viewpoint to discard occluded primitives during 3DGS rendering, combined with an instanced Tensor-Core rasterizer for lower VRAM and better quality on composed scenes.
Characterization and analysis of the 3d gaussian splatting rendering pipeline.IEEE Computer Architecture Letters, 24 (1):13–16
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
NVGS: Neural Visibility for Occlusion Culling in 3D Gaussian Splatting
A shared MLP predicts Gaussian visibility from viewpoint to discard occluded primitives during 3DGS rendering, combined with an instanced Tensor-Core rasterizer for lower VRAM and better quality on composed scenes.