3DGS-LM：使用Levenberg-Marquardt加速高斯飛灑優化

3DGS-LM: Faster Gaussian-Splatting Optimization with Levenberg-Marquardt

September 19, 2024

作者: Lukas Höllein, Aljaž Božič, Michael Zollhöfer, Matthias Nießner

cs.AI

摘要

我們提出了3DGS-LM，這是一種新方法，通過將其ADAM優化器替換為定制的Levenberg-Marquardt（LM）來加速3D高斯擴散（3DGS）的重建。現有方法通過減少高斯數量或改進可微光柵化器的實現來減少優化時間。然而，它們仍然依賴於ADAM優化器來擬合場景中數千次迭代的高斯參數，這可能需要長達一小時的時間。為此，我們將優化器更改為與3DGS可微光柵化器並行運行的LM。為了實現高效的GPU並行處理，我們提出了一種用於中間梯度的緩存數據結構，這使我們能夠在自定義CUDA內核中高效計算雅可比向量乘積。在每個LM迭代中，我們使用這些內核從多個圖像子集計算更新方向，並將它們組合成加權平均值。總的來說，我們的方法比原始的3DGS快30％，同時獲得相同的重建質量。我們的優化方法也不受其他加速3DGS方法的影響，因此與普通3DGS相比，甚至可以實現更快的加速。

English

We present 3DGS-LM, a new method that accelerates the reconstruction of 3D Gaussian Splatting (3DGS) by replacing its ADAM optimizer with a tailored Levenberg-Marquardt (LM). Existing methods reduce the optimization time by decreasing the number of Gaussians or by improving the implementation of the differentiable rasterizer. However, they still rely on the ADAM optimizer to fit Gaussian parameters of a scene in thousands of iterations, which can take up to an hour. To this end, we change the optimizer to LM that runs in conjunction with the 3DGS differentiable rasterizer. For efficient GPU parallization, we propose a caching data structure for intermediate gradients that allows us to efficiently calculate Jacobian-vector products in custom CUDA kernels. In every LM iteration, we calculate update directions from multiple image subsets using these kernels and combine them in a weighted mean. Overall, our method is 30% faster than the original 3DGS while obtaining the same reconstruction quality. Our optimization is also agnostic to other methods that acclerate 3DGS, thus enabling even faster speedups compared to vanilla 3DGS.