SplatQ: Interaction-Aware Pruning of 3D Gaussian Splatting via Quadratic Selection

Key points

• We propose SplatQ, a combinatorial pruning algorithm that coordinates regional budget allocation and joint Gaussian selection under a fixed budget.1
• SplatQ partitions Gaussians by spatial proximity and appearance similarity into size-limited blocks, decomposing global selection into subproblems of a specific size.1
• Experiments on 5 benchmarks demonstrate that SplatQ outperforms state-of-the-art pruning methods in aggregate reconstruction quality, spatial quality balance, and rendering speed.1
• Ablations confirm that budget smoothing improves local detail reconstruction, while interaction modeling improves Gaussian selection quality under a fixed budget.1

Sources (1)

• [1]SplatQ: Interaction-Aware Pruning of 3D Gaussian Splatting via Quadratic Selection
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 9, 12:30 PM

  Existing 3D Gaussian Splatting (3DGS) pruning methods primarily rely on global Top-K selection based on independent importance scores, lacking regional budget control and ignoring joint deletion effects.

  We propose SplatQ, a combinatorial pruning algorithm that coordinates regional budget allocation and joint Gaussian selection under a fixed budget.

Extractive summary: sentences quoted from the sources.