VEDAL: 用于3D高斯泼溅剪枝的变分误差驱动异步学习
VEDAL: Variational Error-Driven Asynchronous Learning for 3D Gaussian Splatting Pruning
AI总结：
提出VEDAL框架，通过变分自由能最小化、预测误差门控机制和变分不确定性头实现3D高斯泼溅的高效剪枝，在5.2倍压缩下仅损失0.31 dB PSNR。
摘要：
3D高斯泼溅（3DGS）通过实时渲染实现了卓越的新视图合成质量，但由于数百万个高斯原语导致内存消耗过大。现有的剪枝方法依赖于启发式重要性分数或同步批量更新，导致压缩次优和训练不稳定。我们提出VEDAL，一个将高斯剪枝公式化为变分自由能最小化的原则性框架。我们的方法引入了（1）一种预测误差门控机制，基于每个高斯的重建不确定性异步激活剪枝，以及（2）一个变分不确定性头，将剪枝决策建模为具有可学习先验的潜变量。自由能目标通过信息论视角自然地平衡了重建保真度与模型复杂度。在Mip-NeRF 360、Tanks&Temples和Deep Blending上的大量实验表明，VEDAL在仅0.31 dB PSNR下降的情况下实现了5.2倍压缩，在更高压缩比下优于PUP 3D-GS 0.05 dB，在相当质量下优于LightGaussian 0.35 dB，同时保持185 FPS的实时渲染。
英文摘要：
3D Gaussian Splatting (3DGS) achieves remarkable novel view synthesis quality with real-time rendering, yet suffers from excessive memory consumption due to millions of Gaussian primitives. Existing pruning methods rely on heuristic importance scores or synchronous batch updates, leading to suboptimal compression and training instability. We propose VEDAL, a principled framework that formulates Gaussian pruning as variational free energy minimization. Our approach introduces (1) a prediction-error gating mechanism that asynchronously activates pruning based on per-Gaussian reconstruction uncertainty, and (2) a variational uncertainty head that models pruning decisions as latent variables with learnable priors. The free energy objective naturally balances reconstruction fidelity against model complexity through an information-theoretic lens. Extensive experiments on Mip-NeRF 360, Tanks&Temples, and Deep Blending demonstrate that VEDAL achieves 5.2x compression with only 0.31 dB PSNR drop, outperforming PUP 3D-GS by +0.05 dB at a higher compression ratio and LightGaussian by +0.35 dB at comparable quality, while maintaining real-time rendering at 185 FPS.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
