GSPotential: Camera Potential Field for Sparse-View 3D Gaussian Splatting

Zeyuan An; Yanghang Xiao; Zhiying Leng; Yijun Feng; Xiaohui Liang

cs.CVAug 29, 2026

GSPotential: Camera Potential Field for Sparse-View 3D Gaussian Splatting

Authors: Zeyuan An, Yanghang Xiao, Zhiying Leng, Yijun Feng, Xiaohui Liang

Organizations: State Key Laboratory of Virtual Reality Technology and Systems, Beihang University, Beijing, China · School of Computer Science and Engineering, Beihang University, Beijing, China · Zhongguancun Laboratory, Beijing, China

Abstract

3D Gaussian Splatting has achieved remarkable success in photorealistic rendering, yet it suffers from severe overfitting and geometric artifacts in sparse-view scenarios due to the inherent deficiency of photometric supervision. Recent advances have attempted to regularize optimization by incorporating external priors, such as depth, point clouds, or diffusion models. However, these methods typically overlook the non-uniform distribution of supervision across the viewing space, resulting in limited specificity in prior use and primitive control. In this paper, we propose GSPotential, a framework that quantifies view-space supervision imbalance using a Camera Potential Field. Our key insight is to identify supervision valleys where photometric constraints are most deficient, and use the potential field to guide reconstruction from two complementary aspects. First, we devise a probabilistic spherical sampling strategy that places informative virtual cameras in low-potential regions. Point-cloud renderings from these views then provide targeted geometric guidance. Second, the same field provides a directional coverage cue for conservative Gaussian updates in weakly covered spatial sectors. Extensive experiments demonstrate that GSPotential achieves high reconstruction fidelity while maintaining competitive training efficiency.

Figures & tables

Figure 1: Visualizations of scene reconstruction (inputs are from GaussianObject [ 30 ] ). Given sparse-view inputs, our GSPotential achieves improved reconstruction quality compared to original 3DGS. The boxes highlight regions with visible improvements.

Figure 2: Overview of GSPotential. Given sparse images, camera poses, and point clouds, we first estimate a Camera Potential Field to model view-space supervision imbalance. Based on this field, we perform Potential-Guided Viewpoint Generation to generate virtual cameras in low-potential regions. The sampled views are then used for Potential-Guided Reconstruction , while the projected potential provides a directional coverage score for Coverage-Aware Primitive Evolution, including anisotropy pruning and conservative densification.

Figure 3: Camera potential fields driven by sampling. We visualize the PSNR of 3DGS rendered images from random viewpoints across view space. (Left) In a single-camera setup, high-quality reconstruction is limited to a narrow cone. (Right) With 4 sparse cameras, despite the increase in coverage, significant supervision valleys (low PSNR zones) persist across viewpoints.

Figure 4: Comparison of kernel choices for modeling camera-potential decay. The horizontal axis $\theta$ denotes the angular distance between a novel viewpoint and its nearest training view. We compare vMF with $\kappa=3.0$ (left), $\mathrm{sech}$ with $\sigma=5.3$ (middle), and vMF with $\kappa=4.4$ (right). The latter two are both calibrated to decay to $1\%$ of their peak at $\theta=90^{\circ}$ , while $\kappa=3.0$ is selected to approximate the overall decay trend of the $\mathrm{sech}$ kernel. Compared with vMF, the $\mathrm{sech}$ kernel yields a smoother relation between camera potential and novel-view rendering quality.

Figure 5: Analysis of camera-potential parameterization and novel-view rendering quality. Each scatter point denotes a sampled viewpoint. The horizontal axis is the camera potential computed with a specific parameter setting, and the vertical axis reports the corresponding novel-view rendering quality. Darker colors indicate larger angular distance to the nearest training view. Compared with using angular distance alone, the potential value provides a more discriminative estimate of supervision strength, since viewpoints with similar nearest-view distance can still exhibit different rendering quality under different camera configurations. This verifies that the camera potential captures the joint camera-layout effect beyond a nearest-angle cue.

Dataset	Mip-NeRF 360				OmniObject3D
Method $\|$ Metric	PSNR $\uparrow$	SSIM $\uparrow$	LPIPS $\downarrow$	TIME/s $\downarrow$	PSNR $\uparrow$	SSIM $\uparrow$	LPIPS $\downarrow$	TIME/s $\downarrow$
3DGS [ 12 ]	13.52	0.282	0.588	500	20.87	0.805	0.221	314
SparseGS [ 28 ]	11.15	0.166	0.672	627	19.15	0.751	0.305	745
FSGS [ 33 ]	13.49	0.324	0.573	762	20.13	0.797	0.206	575
CoR-GS [ 31 ]	12.90	0.309	0.574	2110	21.44	0.813	0.197	1559
SplatFields [ 16 ]	11.92	0.331	0.748	2865	21.12	0.811	0.235	1194

Table 1: Quantitative evaluation of our method compared to previous works on two datasets. Results on Mip-NeRF 360 are averaged over seven scenes, and all metrics are computed on uniformly sampled test views. (“–” denotes unavailable results.)

Figure 6: Comparison of 3DGS training with different point cloud initialization strategies. The plots visualize training iterations (x) versus PSNR (y), with: left for original point cloud (generated by DUSt3R), 1/4, 1/16, and 1/64 down-sampling ratios; right for original point cloud (generated by DUSt3R) and random 3D noise initialization with the same down-sampling ratios.

Figure 7: Qualitative comparison on unsupervised views. Boxes highlight regions with notable improvements by GSPotential.

$\mathcal{L}_{\text{virt}}$	$U_{\text{info}}$	$Q_{i}$ -P	$Q_{i}$ -D	$U_{\text{conf}}$	PSNR $\uparrow$	SSIM $\uparrow$	LPIPS $\downarrow$
$\times$	$\times$	$\times$	$\times$	$\times$	13.52	0.282	0.588
$\checkmark$	$\times$	$\times$	$\times$	$\times$	14.49	0.356	0.624
$\checkmark$	$\checkmark$	$\times$	$\times$	$\times$	15.90	0.379	0.601
$\checkmark$	$\checkmark$	$\checkmark$	$\times$	$\times$	16.03	0.388	0.585
$\checkmark$	$\checkmark$	$\times$	$\checkmark$	$\times$	16.17	0.384	0.594
$\checkmark$	$\checkmark$	$\checkmark$	$\checkmark$	$\times$	16.34	0.394	0.577

Table 2: Ablations of different components of GSPotential, evaluated on Mip-NeRF 360.

U_{\text{info}}

denotes the information potential used for low-potential viewpoint generation.

Q_{i}

-P and

Q_{i}

-D denote coverage-aware pruning and densification guided by the directional coverage score

Q_{i}

, respectively.

U_{\text{conf}}

denotes the confidence term for suppressing unreliable virtual supervision near extrapolation regions.

Figure 8: Qualitative ablation on test view of the Mip-NeRF 360 counter scene. Columns correspond to the variants in Table 2 .

Views	Ours			3DGS
Views	PSNR $\uparrow$	SSIM $\uparrow$	LPIPS $\downarrow$	PSNR $\uparrow$	SSIM $\uparrow$	LPIPS $\downarrow$
1	15.73	0.392	0.571	13.02	0.281	0.604
2	16.14	0.408	0.527	13.21	0.282	0.597
4	16.43	0.411	0.505	13.52	0.282	0.588
8	16.59	0.415	0.480	13.78	0.282	0.581

Table 3: Ablation of point cloud quality, evaluated on Mip-NeRF 360. Views refer to number of views used to build the point cloud.

Sampling	PSNR $\uparrow$	SSIM $\uparrow$	LPIPS $\downarrow$
Nearest-angle random	15.63	0.318	0.520
Uniform sphere	15.58	0.306	0.525
Farthest-point sphere	15.50	0.305	0.539
Uniform interpolation	15.52	0.310	0.522
CPF (ours)	15.93	0.373	0.468

Table 4: Controlled comparison of virtual-view sampling strategies on the Mip-NeRF 360 kitchen scene. All variants use the same reconstruction setting and differ only in virtual-view selection.

Views	3DGS	GSPotential
3	14.81/0.331/0.508	15.49/0.358/0.478
4	15.09/0.337/0.495	15.93/0.373/0.468
5	15.78/0.398/0.460	17.00/0.412/0.441
6	15.59/0.359/0.495	15.89/0.413/0.464
9	16.81/0.469/0.441	17.19/0.498/0.404

Table 5: Sensitivity to the number of input views on the Mip-NeRF 360 kitchen scene. Each entry reports PSNR/SSIM/LPIPS.

Explore similar work

SparseGS: Sparse View Synthesis using 3D Gaussian Splatting
Nov 30, 2023Haolin Xiong, Sairisheek Muttukuru, Hanyuan Xiao +43D Gaussian Splatting Novel View Synthesis
University of California, Los Angeles · USC Institute for Creative Technologies · University of Southern California
AugSplat: Radiance Field-Informed Gaussian Splatting for Sparse-View Settings
Jun 30, 2026Lorenzo Lazzaroni, Riccardo Bollati, Daniel Barath +23D Gaussian Splatting Synthetic Data Augmentation
ETH Zurich · 1Google · 2ETH Zurich
Improving Sparse-View 3DGS Generalization via Flat Minima Optimization
Jul 1, 2026Kangmin Seo, Sangeek Hyun, MinKyu Lee +13D Gaussian Splatting Novel View Synthesis
Sungkyunkwan University, South Korea

Nov 30, 2023J/K move · Enter open · S save

SparseGS: Sparse View Synthesis using 3D Gaussian Splatting

Haolin Xiong, Sairisheek Muttukuru, Hanyuan Xiao, Rishi Upadhyay+3

University of California, Los Angeles · USC Institute for Creative Technologies · University of Southern California

3D Gaussian Splatting Novel View Synthesis Sparse-View 3D Reconstruction

Open paper AskPDF arXiv

3D Gaussian Splatting (3DGS) has recently enabled real-time rendering of unbounded 3D scenes for novel view synthesis. However, this technique requires dense training views to accurately reconstruct 3D geometry. A limited number of input views will significantly degrade reconstruction quality, resulting in artifacts such as "floaters" and "background collapse" at unseen viewpoints. In this work, we introduce SparseGS, an efficient training pipeline designed to address the limitations of 3DGS in scenarios with sparse training views. SparseGS incorporates depth priors, novel depth rendering techniques, and a pruning heuristic to mitigate floater artifacts, alongside an Unseen Viewpoint Regularization module to alleviate background collapses. Our extensive evaluations on the Mip-NeRF360, LLFF, and DTU datasets demonstrate that SparseGS achieves high-quality reconstruction in both unbounded and forward-facing scenarios, with as few as 12 and 3 input images, respectively, while maintaining fast training and real-time rendering capabilities.

Abstract

Explore similar work

SparseGS: Sparse View Synthesis using 3D Gaussian Splatting

AugSplat: Radiance Field-Informed Gaussian Splatting for Sparse-View Settings

Improving Sparse-View 3DGS Generalization via Flat Minima Optimization