The Role and Relationship of Initialization and Densification in 3D Gaussian Splatting

• URL: http://arxiv.org/abs/2603.20714v1
• Date: Sat, 21 Mar 2026 08:37:36 GMT
• Title: The Role and Relationship of Initialization and Densification in 3D Gaussian Splatting
• Abstract summary: 3DGS represents the scene through a set of 3D Gaussians, parameterized by their position, spatial extent, and view-dependent color.<n>Current densification approaches are not able to take full advantage of dense.<n>introductors as they are often unable to (significantly) improve over sparse.<n>SfM point clouds.<n>We show that current densification approaches are not able to take full advantage of dense.<n>introductors as they are often unable to (significantly) improve over sparse.<n>SfM point clouds.
• Score: 23.100379811878867
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 3D Gaussian Splatting (3DGS) has become the method of choice for photo-realistic 3D reconstruction of scenes, due to being able to efficiently and accurately recover the scene appearance and geometry from images. 3DGS represents the scene through a set of 3D Gaussians, parameterized by their position, spatial extent, and view-dependent color. Starting from an initial point cloud, 3DGS refines the Gaussians' parameters as to reconstruct a set of training images as accurately as possible. Typically, a sparse Structure-from-Motion point cloud is used as initialization. In order to obtain dense Gaussian clouds, 3DGS methods thus rely on a densification stage. In this paper, we systematically study the relation between densification and initialization. Proposing a new benchmark, we study combinations of different types of initializations (dense laser scans, dense (multi-view) stereo point clouds, dense monocular depth estimates, sparse SfM point clouds) and different densification schemes. We show that current densification approaches are not able to take full advantage of dense initialization as they are often unable to (significantly) improve over sparse SfM-based initialization. We will make our benchmark publicly available.

Related papers

• Initialize to Generalize: A Stronger Initialization Pipeline for Sparse-View 3DGS [31.824354716471294]
  Sparse-view 3D Gaussian Splatting (3DGS) often overfits to the training views, leading to artifacts like blurring in novel view rendering.<n>Prior work addresses it either by enhancing the point cloud from Structure-from-Motion (SfM) or by adding training-time constraints (regularization) to the 3DGS optimization.<n>We design frequency-aware SfM that improves low-texture coverage via low-frequency view augmentation and relaxed multi-view correspondences.
  arXiv  Detail & Related papers  (2025-10-20T12:23:19Z)
• Sparse2DGS: Sparse-View Surface Reconstruction using 2D Gaussian Splatting with Dense Point Cloud [9.784526657786342]
  We propose a new 3D reconstruction method, called Sparse2DGS, to enhance 2DGS in reconstructing objects using only three images.<n>Sparse2DGS employs DUSt3R, a fundamental model for stereo images, along with COLMAP MVS to generate highly accurate and dense 3D point clouds.<n>We show that Sparse2DGS can accurately reconstruct the 3D shapes of objects using just three images.
  arXiv  Detail & Related papers  (2025-05-26T11:38:26Z)
• FreeSplat++: Generalizable 3D Gaussian Splatting for Efficient Indoor Scene Reconstruction [50.534213038479926]
  FreeSplat++ is an alternative approach to large-scale indoor whole-scene reconstruction.<n>Our method with depth-regularized per-scene fine-tuning demonstrates substantial improvements in reconstruction accuracy and a notable reduction in training time.
  arXiv  Detail & Related papers  (2025-03-29T06:22:08Z)
• EasySplat: View-Adaptive Learning makes 3D Gaussian Splatting Easy [34.27245715540978]
  We introduce a novel framework EasySplat to achieve high-quality 3DGS modeling.<n>We propose an efficient grouping strategy based on view similarity, and use robust pointmap priors to obtain high-quality point clouds.<n>After obtaining a reliable scene structure, we propose a novel densification approach that adaptively splits Gaussian primitives based on the average shape of neighboring Gaussian ellipsoids.
  arXiv  Detail & Related papers  (2025-01-02T01:56:58Z)
• No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images [100.80376573969045]
  NoPoSplat is a feed-forward model capable of reconstructing 3D scenes parameterized by 3D Gaussians from multi-view images. Our model achieves real-time 3D Gaussian reconstruction during inference. This work makes significant advances in pose-free generalizable 3D reconstruction and demonstrates its applicability to real-world scenarios.
  arXiv  Detail & Related papers  (2024-10-31T17:58:22Z)
• PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting [54.7468067660037]
  PF3plat sets a new state-of-the-art across all benchmarks, supported by comprehensive ablation studies validating our design choices.<n>Our framework capitalizes on fast speed, scalability, and high-quality 3D reconstruction and view synthesis capabilities of 3DGS.
  arXiv  Detail & Related papers  (2024-10-29T15:28:15Z)
• LP-3DGS: Learning to Prune 3D Gaussian Splatting [71.97762528812187]
  We propose learning-to-prune 3DGS, where a trainable binary mask is applied to the importance score that can find optimal pruning ratio automatically. Experiments have shown that LP-3DGS consistently produces a good balance that is both efficient and high quality.
  arXiv  Detail & Related papers  (2024-05-29T05:58:34Z)
• InstantSplat: Sparse-view Gaussian Splatting in Seconds [91.77050739918037]
  We introduce InstantSplat, a novel approach for addressing sparse-view 3D scene reconstruction at lightning-fast speed.<n>InstantSplat employs a self-supervised framework that optimize 3D scene representation and camera poses.<n>It achieves an acceleration of over 30x in reconstruction and improves visual quality (SSIM) from 0.3755 to 0.7624 compared to traditional SfM with 3D-GS.
  arXiv  Detail & Related papers  (2024-03-29T17:29:58Z)
• Relaxing Accurate Initialization Constraint for 3D Gaussian Splatting [29.58220473268378]
  We propose a novel optimization strategy dubbed RAIN-GS (Relaing Accurate Initialization Constraint for 3D Gaussian Splatting) RAIN-GS successfully trains 3D Gaussians from sub-optimal point cloud (e.g., randomly point cloud) We demonstrate the efficacy of our strategy through quantitative and qualitative comparisons on multiple datasets, where RAIN-GS trained with random point cloud achieves performance on-par with or even better than 3DGS trained with accurate SfM point cloud.
  arXiv  Detail & Related papers  (2024-03-14T14:04:21Z)
• GaussianPro: 3D Gaussian Splatting with Progressive Propagation [49.918797726059545]
  3DGS relies heavily on the point cloud produced by Structure-from-Motion (SfM) techniques. We propose a novel method that applies a progressive propagation strategy to guide the densification of the 3D Gaussians. Our method significantly surpasses 3DGS on the dataset, exhibiting an improvement of 1.15dB in terms of PSNR.
  arXiv  Detail & Related papers  (2024-02-22T16:00:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.