EDGS：消除致密化以实现 3DGS 的高效收敛
EDGS: Eliminating Densification for Efficient Convergence of 3DGS
用稠密图像对应三角化得到的像素点云替代 SfM 稀疏点做 3DGS 初始化，直接省掉 densification 阶段；做 3DGS 加速或高质量重建时，可把这套稠密初始化当作即插即用的起点，与其它加速手段叠加。
原文依据 eliminate densification with a one-step approximation of scene geometry using triangulated pixels from dense image correspondences
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论3D 高斯泼溅从稀疏的运动恢复结构初始化出发重建场景，并对重建不足的区域进行细化。该过程较慢，因为它需要多个致密化步骤，在这些步骤中高斯被反复分裂与调整，并沿着一条漫长的优化路径进行。此外，这种增量式方法在高频区域往往产生次优的渲染结果。我们提出一种根本不同的思路：利用来自稠密图像对应关系的三角化像素，以一步式场景几何近似来消除致密化。这种稠密初始化使我们能够估计场景的粗略几何，同时保留输入 RGB 图像中的丰富细节，为每个高斯提供信息充分的颜色、尺度与位置。因此，我们大幅缩短了优化路径，并消除了对致密化的需求。与依赖稀疏关键点的方法不同，我们的稠密初始化确保整个场景的细节均匀，即使在其他方法难以处理的高频区域也是如此。此外，由于所有泼溅在优化开始时即并行初始化，我们无需等待致密化来调整新的高斯。EDGS 达到标准 3DGS 的感知相似度与结构相似度性能的速度显著快于现有的以效率为目标的方法。在进一步训练后，其重建质量超过以最大化保真度为目标的最优模型。我们的方法与其他加速技术完全兼容，是一种通用且高效的解决方案，可与现有方法集成。
英文原文摘要 1600 字符
3D Gaussian Splatting reconstructs scenes by starting from a sparse Structure-from-Motion initialization and refiningunder-reconstructed regions. This process is slow, as it requires multiple densification steps where Gaussians arerepeatedly split and adjusted, following a lengthy optimization path. Moreover, this incremental approach often yieldssuboptimal renderings in high-frequency regions. We propose a fundamentally different approach: eliminate densification with a one-step approximation of scenegeometry using triangulated pixels from dense image correspondences. This dense initialization allows us to estimatethe rough geometry of the scene while preserving rich details from input RGB images, providing each Gaussian withwell-informed color, scale, and position. As a result, we dramatically shorten the optimization path and remove theneed for densification. Unlike methods that rely on sparse keypoints, our dense initialization ensures uniform detailacross the scene, even in high-frequency regions where other methods struggle. Moreover, since all splats are initializedin parallel at the start of optimization, we remove the need to wait for densification to adjust new Gaussians.EDGS reaches LPIPS and SSIM performance of standard 3DGS significantly faster than existing efficiency-focused approaches. When trained further, it exceeds the reconstruction quality of state-of-the-art models aimed at maximizing fidelity. Our method is fully compatible with other acceleration techniques, making it a versatile and efficient solution that can be integrated with existing approaches.