AD-GS: Alternating Densification for Sparse-Input 3D Gaussian Splatting

SIGGRAPH Asia 2025
1Indian Institute of Science (IISc), Bengaluru

AD-GS renders the artifact free novel views from sparse input views.

Abstract

3D Gaussian Splatting (3DGS) has shown impressive results in real-time novel view synthesis. However, it often struggles under sparse-view settings, producing undesirable artifacts such as floaters, inaccurate geometry, and overfitting due to limited observations.

We propose AD-GS, a novel alternating densification framework that interleaves high and low densification phases. During high densification, the model densifies aggressively, followed by photometric loss based training to capture fine-grained scene details. Low densification then primarily involves aggressive opacity pruning of Gaussians followed by regularizing their geometry through pseudo-view consistency and edge-aware depth smoothness.

This alternating approach helps reduce overfitting by carefully controlling model capacity growth while progressively refining the scene representation. Extensive experiments on challenging datasets demonstrate that AD-GS significantly improves rendering quality and geometric consistency compared to existing methods.

Method

AD-GS Method: The training begins with a Warm-Up Phase of NW iterations, where two 3DGS models are trained independently using photometric loss. This is followed by the Alternating Densification Phase, which alternates between low and high densification steps every NL and NH iterations, respectively. Low Densification applies strict gradient and opacity thresholds and includes additional supervision via pseudo-view consistency and edge-aware depth smoothness losses. The total loss used during this phase is defined as the "Combined Loss". High Densification enables aggressive Gaussian growth and uses only photometric loss to recover high-frequency details. Over iterations, this alternation progressively refines scene geometry while avoiding overfitting and floaters. The figure illustrates how Gaussian counts increase during high densification and decrease due to pruning during low densification.

3DGS vs AD-GS teaser image

Visuals

Drag and Check Novel frames

Below we can check the novel view renders. Use the slider here.

Loading...

Observations

Challenges of Densification in Sparse View Training

As training progresses in 3DGS, densification under sparse views leads to floating artifacts and inconsistent geometry.

3DGS vs AD-GS teaser image

Floaters Suppression with AD-GS

3DGS develops floaters (as seen in the red boxes) as we train for more iterations due to its uncontrolled densification. However, our AD-GS model is able to suppress floaters.

3DGS vs AD-GS teaser image

Detail Preservation with AD-GS

CoR-GS tends to employ aggressive smoothing to resolve floaters at the cost of loss of details, whereas AD-GS is able to recon- struct textures better.

3DGS vs AD-GS teaser image

Comparison with other methods

3DGS vs AD-GS

3DGS tends to produce many floaters and noisy artifacts in the reconstruction, especially under sparse input conditions. In contrast, AD-GS resolves these issues by delivering cleaner and more stable reconstructions with minimal floaters.

FSGS vs AD-GS

FSGS often suffers from poor reconstruction quality with visible artifacts and distortions in the output. AD-GS overcomes these limitations by ensuring high fidelity and stable reconstructions, even with sparse input data.

CoR-GS vs AD-GS

CoR-GS results in loss of fine details and produces overly smooth regions that degrade reconstruction quality. AD-GS effectively preserves details while maintaining stability and sharpness in the reconstructed scenes.

DropGaussian vs AD-GS

DropGaussian suffers from floaters and overall less accurate reconstructions, with noticeable artifacts present. AD-GS successfully addresses these challenges, providing more accurate and artifact-free reconstructions.

More Visual Results

For more visual results, please check here.

Acknowledgement

This work was supported in part by the Kotak IISc AI–ML Centre (KIAC).

Citation

If you find our code or paper useful, please cite:

@inproceedings{Patle2025ADGS,
 author = {Patle, Gurutva and Girgaonkar, Nilay and Somraj, Nagabhushan and Soundararajan, Rajiv},
 title = {{AD-GS: Alternating Densification for Sparse-Input 3D Gaussian Splatting}},
 booktitle = {{Proceedings of SIGGRAPH Asia 2025 Conference Papers}},
 series = {{SA '25}},
 year = {2025},
 note = {Accepted, In Press},
 publisher = {{ACM}},
 address = {Hong Kong},
 doi = {10.1145/3757377.3763993},
 eprint = {2509.11003},
 archivePrefix = {arXiv},
 primaryClass = {cs.GR}
}
@article{patle2025ad,
  title   = {AD-GS: Alternating Densification for Sparse-Input 3D Gaussian Splatting},
  author  = {Patle, Gurutva and Girgaonkar, Nilay and Somraj, Nagabhushan and Soundararajan, Rajiv},
  journal = {arXiv preprint arXiv:2509.11003},
  year    = {2025}
}