DropGaussian: Structural Regularization for Sparse-view Gaussian Splatting

• URL: http://arxiv.org/abs/2504.00773v1
• Date: Tue, 01 Apr 2025 13:23:34 GMT
• Title: DropGaussian: Structural Regularization for Sparse-view Gaussian Splatting
• Authors: Hyunwoo Park, Gun Ryu, Wonjun Kim,
• Abstract summary: This paper introduces a prior-free method, so-called DropGaussian, with simple changes in 3D Gaussian splatting.<n>Specifically, we randomly remove Gaussians during the training process in a similar way of dropout, which allows non-excluded Gaussians to have larger gradients.<n>Such simple operation effectively alleviates the overfitting problem and enhances the quality of novel view synthesis.
• Score: 5.216151302783165
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recently, 3D Gaussian splatting (3DGS) has gained considerable attentions in the field of novel view synthesis due to its fast performance while yielding the excellent image quality. However, 3DGS in sparse-view settings (e.g., three-view inputs) often faces with the problem of overfitting to training views, which significantly drops the visual quality of novel view images. Many existing approaches have tackled this issue by using strong priors, such as 2D generative contextual information and external depth signals. In contrast, this paper introduces a prior-free method, so-called DropGaussian, with simple changes in 3D Gaussian splatting. Specifically, we randomly remove Gaussians during the training process in a similar way of dropout, which allows non-excluded Gaussians to have larger gradients while improving their visibility. This makes the remaining Gaussians to contribute more to the optimization process for rendering with sparse input views. Such simple operation effectively alleviates the overfitting problem and enhances the quality of novel view synthesis. By simply applying DropGaussian to the original 3DGS framework, we can achieve the competitive performance with existing prior-based 3DGS methods in sparse-view settings of benchmark datasets without any additional complexity. The code and model are publicly available at: https://github.com/DCVL-3D/DropGaussian release.

Related papers

• VGNC: Reducing the Overfitting of Sparse-view 3DGS via Validation-guided Gaussian Number Control [25.771146052678368]
  We introduce Validation-guided Gaussian Number Control (VGNC) approach based on generative novel view synthesis (NVS) models. This is the first attempt to alleviate the overfitting issue of sparse-view 3DGS with generative validation images.
  arXiv  Detail & Related papers  (2025-04-20T09:38:02Z)
• DropoutGS: Dropping Out Gaussians for Better Sparse-view Rendering [45.785618745095164]
  3D Gaussian Splatting (3DGS) has demonstrated promising results in novel view synthesis. As the number of training views decreases, the novel view synthesis task degrades to a highly under-determined problem. We propose a Random Dropout Regularization (RDR) to exploit the advantages of low-complexity models to alleviate overfitting. In addition, to remedy the lack of high-frequency details for these models, an Edge-guided Splitting Strategy (ESS) is developed.
  arXiv  Detail & Related papers  (2025-04-13T09:17:21Z)
• NovelGS: Consistent Novel-view Denoising via Large Gaussian Reconstruction Model [57.92709692193132]
  NovelGS is a diffusion model for Gaussian Splatting given sparse-view images. We leverage the novel view denoising through a transformer-based network to generate 3D Gaussians.
  arXiv  Detail & Related papers  (2024-11-25T07:57:17Z)
• PixelGaussian: Generalizable 3D Gaussian Reconstruction from Arbitrary Views [116.10577967146762]
  PixelGaussian is an efficient framework for learning generalizable 3D Gaussian reconstruction from arbitrary views. Our method achieves state-of-the-art performance with good generalization to various numbers of views.
  arXiv  Detail & Related papers  (2024-10-24T17:59:58Z)
• Effective Rank Analysis and Regularization for Enhanced 3D Gaussian Splatting [33.01987451251659]
  3D Gaussian Splatting (3DGS) has emerged as a promising technique capable of real-time rendering with high-quality 3D reconstruction.<n>Despite its potential, 3DGS encounters challenges such as needle-like artifacts, suboptimal geometries, and inaccurate normals.<n>We introduce the effective rank as a regularization, which constrains the structure of the Gaussians.
  arXiv  Detail & Related papers  (2024-06-17T15:51:59Z)
• DOGS: Distributed-Oriented Gaussian Splatting for Large-Scale 3D Reconstruction Via Gaussian Consensus [56.45194233357833]
  We propose DoGaussian, a method that trains 3DGS distributedly. Our method accelerates the training of 3DGS by 6+ times when evaluated on large-scale scenes.
  arXiv  Detail & Related papers  (2024-05-22T19:17:58Z)
• AbsGS: Recovering Fine Details for 3D Gaussian Splatting [10.458776364195796]
  3D Gaussian Splatting (3D-GS) technique couples 3D primitives with differentiable Gaussianization to achieve high-quality novel view results. However, 3D-GS frequently suffers from over-reconstruction issue in intricate scenes containing high-frequency details, leading to blurry rendered images. We present a comprehensive analysis of the cause of aforementioned artifacts, namely gradient collision. Our strategy efficiently identifies large Gaussians in over-reconstructed regions, and recovers fine details by splitting.
  arXiv  Detail & Related papers  (2024-04-16T11:44:12Z)
• CoherentGS: Sparse Novel View Synthesis with Coherent 3D Gaussians [18.42203035154126]
  We introduce a structured Gaussian representation that can be controlled in 2D image space.<n>We then constraint the Gaussians, in particular their position, and prevent them from moving independently during optimization.<n>We demonstrate significant improvements compared to the state-of-the-art sparse-view NeRF-based approaches on a variety of scenes.
  arXiv  Detail & Related papers  (2024-03-28T15:27:13Z)
• 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)
• Spec-Gaussian: Anisotropic View-Dependent Appearance for 3D Gaussian Splatting [55.71424195454963]
  Spec-Gaussian is an approach that utilizes an anisotropic spherical Gaussian appearance field instead of spherical harmonics. Our experimental results demonstrate that our method surpasses existing approaches in terms of rendering quality. This improvement extends the applicability of 3D GS to handle intricate scenarios with specular and anisotropic surfaces.
  arXiv  Detail & Related papers  (2024-02-24T17:22:15Z)
• 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)
• GaussianObject: High-Quality 3D Object Reconstruction from Four Views with Gaussian Splatting [82.29476781526752]
  Reconstructing and rendering 3D objects from highly sparse views is of critical importance for promoting applications of 3D vision techniques. GaussianObject is a framework to represent and render the 3D object with Gaussian splatting that achieves high rendering quality with only 4 input images. GaussianObject is evaluated on several challenging datasets, including MipNeRF360, OmniObject3D, OpenIllumination, and our-collected unposed images.
  arXiv  Detail & Related papers  (2024-02-15T18:42:33Z)
• GES: Generalized Exponential Splatting for Efficient Radiance Field Rendering [112.16239342037714]
  GES (Generalized Exponential Splatting) is a novel representation that employs Generalized Exponential Function (GEF) to model 3D scenes. With the aid of a frequency-modulated loss, GES achieves competitive performance in novel-view synthesis benchmarks.
  arXiv  Detail & Related papers  (2024-02-15T17:32:50Z)

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.