Pith. sign in

REVIEW 8 cited by

DNGaussian: Optimizing Sparse-View 3D Gaussian Radiance Fields with Global-Local Depth Normalization

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2403.06912 v3 pith:NAVM5QIV submitted 2024-03-11 cs.CV

Jiahe Li , Jiawei Zhang , Xiao Bai , Jin Zheng , Xin Ning , Jun Zhou , Lin Gu This is my paper · ORCID
classification cs.CV
keywords depthgaussianfieldsgeometryradiancedngaussianviewscosts
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
abstract

Radiance fields have demonstrated impressive performance in synthesizing novel views from sparse input views, yet prevailing methods suffer from high training costs and slow inference speed. This paper introduces DNGaussian, a depth-regularized framework based on 3D Gaussian radiance fields, offering real-time and high-quality few-shot novel view synthesis at low costs. Our motivation stems from the highly efficient representation and surprising quality of the recent 3D Gaussian Splatting, despite it will encounter a geometry degradation when input views decrease. In the Gaussian radiance fields, we find this degradation in scene geometry primarily lined to the positioning of Gaussian primitives and can be mitigated by depth constraint. Consequently, we propose a Hard and Soft Depth Regularization to restore accurate scene geometry under coarse monocular depth supervision while maintaining a fine-grained color appearance. To further refine detailed geometry reshaping, we introduce Global-Local Depth Normalization, enhancing the focus on small local depth changes. Extensive experiments on LLFF, DTU, and Blender datasets demonstrate that DNGaussian outperforms state-of-the-art methods, achieving comparable or better results with significantly reduced memory cost, a $25 \times$ reduction in training time, and over $3000 \times$ faster rendering speed.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Flow Distillation Sampling: Regularizing 3D Gaussians with Pre-trained Matching Priors

    cs.CV 2025-02 conditional novelty 6.0 of 10

    A new loss that matches an optical-flow model's predictions against analytically computed flows from 3D Gaussians, improving geometric reconstruction on sparse indoor scenes.

  2. sshELF: Single-Shot Hierarchical Extrapolation of Latent Features for 3D Reconstruction from Sparse-Views

    cs.CV 2025-02 conditional novelty 6.0 of 10

    sshELF reconstructs full 360-degree outdoor scenes from six sparse views in 0.18 seconds by generating intermediate virtual views before decoding 3D Gaussian primitives.

  3. RDG-GS: Relative Depth Guidance with Gaussian Splatting for Real-time Sparse-View 3D Rendering

    cs.CV 2025-01 conditional novelty 6.0 of 10

    RDG-GS combines refined monocular depth priors, a relative depth similarity loss, and adaptive point densification to improve sparse-view 3D Gaussian Splatting rendering.

  4. 4D Gaussian Splatting with Scale-aware Residual Field and Adaptive Optimization for Real-time Rendering of Temporally Complex Dynamic Scenes

    cs.CV 2024-12 conditional novelty 6.0 of 10

    SaRO-GS models dynamic scenes with 4D Gaussians plus a scale-aware residual field and adaptive per-Gaussian optimization, achieving state-of-the-art PSNR at real-time frame rates on D-NeRF and Plenoptic Video datasets.

  5. SparseLGS: Sparse View Language Embedded Gaussian Splatting

    cs.CV 2024-12 conditional novelty 6.0 of 10

    SparseLGS builds open-vocabulary 3D semantic fields from 3-4 pose-free images, using learning-based pose/point-cloud estimation and a three-step multi-view semantic alignment.

  6. PhysMotion: Physics-Grounded Dynamics From a Single Image

    cs.CV 2024-11 conditional novelty 6.0 of 10

    PhysMotion generates physically plausible videos from a single image by simulating 3D object motion with a material point method, then enhancing the rendering with a diffusion model.

  7. FatesGS: Fast and Accurate Sparse-View Surface Reconstruction using Gaussian Splatting with Depth-Feature Consistency

    cs.CV 2025-01 conditional novelty 5.0 of 10

    FatesGS combines local monocular depth ranking, depth smoothing, and multi-view feature alignment in a 2D Gaussian splatting pipeline to obtain accurate surface meshes from only three views without dataset-scale pre-training.

  8. GBR: Generative Bundle Refinement for High-fidelity Gaussian Splatting with Enhanced Mesh Reconstruction

    cs.CV 2024-12 conditional novelty 5.0 of 10

    GBR reconstructs accurate camera poses, dense point clouds, and high-fidelity meshes from 4-6 unposed images by combining DUSt3R-based neural bundle adjustment with scale-preserving diffusion depth refinement.

Pith tools