PointSplat: Efficient Geometry-Driven Pruning and Transformer Refinement for 3D Gaussian Splatting
This work addresses memory and storage demands in 3DGS for novel view synthesis, offering an efficient pruning method that eliminates reliance on 2D images during pruning.
PointSplat introduces a geometry-driven pruning and transformer refinement framework for 3D Gaussian Splatting that reduces model size without per-scene optimization, achieving competitive rendering quality and superior efficiency on ScanNet++ and Replica datasets.
3D Gaussian Splatting (3DGS) has recently unlocked real-time, high-fidelity novel view synthesis by representing scenes using explicit 3D primitives. However, traditional methods often require millions of Gaussians to capture complex scenes, leading to significant memory and storage demands. Recent approaches have addressed this issue through pruning and per-scene fine-tuning of Gaussian parameters, thereby reducing the model size while maintaining visual quality. These strategies typically rely on 2D images to compute important scores followed by scene-specific optimization. In this work, we introduce PointSplat, 3D geometry-driven prune-and-refine framework that bridges previously disjoint directions of gaussian pruning and transformer refinement. Our method includes two key components: (1) an efficient geometry-driven strategy that ranks Gaussians based solely on their 3D attributes, removing reliance on 2D images during pruning stage, and (2) a dual-branch encoder that separates, re-weights geometric and appearance to avoid feature imbalance. Extensive experiments on ScanNet++ and Replica across varying sparsity levels demonstrate that PointSplat consistently achieves competitive rendering quality and superior efficiency without additional per-scene optimization.