Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

AAA-Gaussians: Anti-Aliased and Artifact-Free 3D Gaussian Rendering

T0 review · 3 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read 3D Gaussian rendering goes artifact-free at 100+ FPS

desk verdict A solid, well-derived 3DGS rendering paper whose 'exact' culling claim is the one real gap between the math and the text. read the letter →

arxiv 2504.12811 v2 pith:JYPGA3IY submitted 2025-04-17 cs.GR cs.CV

Michael Steiner , Thomas Köhler , Lukas Radl , Felix Windisch , Dieter Schmalstieg , Markus Steinberger This is my paper · ORCID
classification cs.GRcs.CV
keywords 3DGaussianSplattinganti-aliasingview-consistentrenderingevaluationfrustumcullinghierarchicalrasterizationout-of-distributionviewsreal-time
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

3D Gaussian Splatting renders scenes by projecting 3D Gaussians to 2D splats, and this paper argues that those projections are the root of the artifacts that appear when a trained scene is viewed from unusual angles, distances, or fields of view. AAA-Gaussians instead evaluates each Gaussian in full 3D throughout the entire rasterization pipeline, replacing the 2D splat approximation with a maximum-contribution-along-the-ray model. On top of that it adds an adaptive 3D smoothing filter whose amplitude scaling depends only on the Gaussian's area perpendicular to the view ray, so zooming out and moving close no longer cause flicker or over-transparency. The result is a rasterizer that matches the best published quality on standard test views while sharply outperforming them on out-of-distribution views, at frame times of 4-11 ms on a consumer GPU.

What carries the argument

The load-bearing object is the 3D Gaussian evaluated along a viewing ray at its point of maximum contribution, inherited from hybrid transparency rendering via screen-space planes. Three new mechanisms carry the argument: (1) an adaptive 3D smoothing filter that combines the training-time sampling frequency $\hat v_{\text{train}}$ with the current view's frequency $\hat v' = \min(\hat v_{\text{train}}, \hat v)$ and renormalizes each Gaussian by the perpendicular covariance determinant, which is what removes aliasing without over-transparency; (2) perspective-correct bounding that computes tangent angles $\theta_{1,2}, \phi_{1,2}$ of the cutoff ellipsoid in view space and clamps them to $[-\pi/2+\epsilon, \pi/2-\epsilon]$, which is what prevents popping for Gaussians crossing the image plane; and (3) frustum-based 3D tile culling, which constructs per-tile frusta and keeps a tile only if the minimum $\rho(x)^2$ over the frustum is below the opacity threshold $\tau_\rho$, which is what makes hierarchical sorting fast enough for real time.

What would settle it

Render a scene of overlapping semi-transparent anisotropic Gaussians with full volumetric ray tracing as ground truth and with this rasterizer at the same out-of-distribution poses; if the images differ in edge opacity or color by more than a small threshold, the maximum-along-ray model that the filter is built on is falsified. A cheaper check on a single Gaussian: render one highly elongated ellipsoid edge-on at two zoom levels and verify that its opacity stays constant under the perpendicular-area normalization, which is what Eq. (10) predicts.

Watch

Extended reading notes

Core claim

The paper's central claim is that one rasterization pipeline can eliminate all three classic 3DGS artifacts—aliasing, popping, and projection distortion—at once by treating Gaussians as true 3D objects rather than 2D splats. Concretely, the paper replaces the 2D screen-space Mip filter with a 3D smoothing filter that dilates each Gaussian by the training sampling frequency and renormalizes amplitude using only the covariance projected onto the plane perpendicular to the ray, giving a closed-form scale factor $\sqrt{|\Sigma|\,d^\top\Sigma^{-1}d/|\hat\Sigma|\,d^\top\hat\Sigma^{-1}d}$. It bounds Gaussians in view space by solving for the tangent angles of their cutoff ellipsoid, so Gaussians reaching behind the near plane are bounded stably instead of being discarded, and it lifts tile-based culling into 3D by constructing per-tile frusta and discarding tiles where the minimum $\rho(x)^2$ inside the frustum exceeds the threshold. With these components the method reports state-of-the-art or matching metrics on Mip-NeRF 360, Tanks & Temples, and Deep Blending, large improvement over all baselines at 3x larger field of view, and effectively unchanged quality at half, full, and double resolution.

Load-bearing premise

The whole anti-aliasing calibration rests on the approximation that a Gaussian's contribution to a pixel is its maximum value along the ray, not the integral of density along that ray; if true volumetric integration is required for correctness, the new perpendicular amplitude scaling would be miscalibrated.

Editorial extensions

If this is right

  • Rendering a trained scene at resolutions and fields of view far outside the training distribution no longer degrades: the large-FOV experiment holds PSNR essentially constant while all compared methods drop by several dB.
  • Zooming toward an object or pulling back no longer produces popping or flicker, because the 3D filter and view-space bounds keep contribution stable as the camera moves.
  • The pipeline keeps the MCMC densification training setup, so artifact-free rendering does not require a new representation or a new optimizer.
  • Timing measurements place the full method at 5.8-10.7 ms per frame, i.e. above 100 FPS on an RTX 4090, with culling recovering most of the cost of per-pixel sorting.
  • By removing view-inconsistent 'cheating' in the optimizer, the method exposes the view-dependent color encoding as the next bottleneck, so more expressive encodings should yield disproportionately larger gains.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The perpendicular-area normalization is not specific to Gaussian splatting: any renderer that shades a primitive by its brightest point along a ray could reuse the same scale factor, so the formula is a candidate building block for other ray-max approximations.
  • The view-space bounding construction, although derived for a pinhole camera, is expressed in angles around the camera origin rather than screen coordinates; adapting it to fisheye or other central cameras may be straightforward, and would let the no-popping guarantee carry over to those models.
  • If future work moves to full volumetric integration along the ray, the max-along-ray approximation behind the filter calibration would need to be revisited: the paper's own derivation assumes only the maximum contribution matters, so a testable prediction is that scenes with many overlapping translucent Gaussians will be the first place the amplitude calibration deviates from a true volume render.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper presents AAA-Gaussians, a rasterization-based 3D Gaussian renderer that evaluates Gaussians in 3D throughout the pipeline. Three contributions are proposed: an adaptive 3D smoothing filter that dilates Gaussians based on the area perpendicular to the viewing ray, a view-space angle bounding method that stabilizes Gaussians extending behind the image plane, and frustum-based 3D tile culling for hierarchical sorting. The authors evaluate on Mip-NeRF 360, Tanks & Temples, and Deep Blending, reporting large-FOV, multi-resolution, and timing results, and argue that their method removes aliasing, popping, and projection distortions while remaining real-time.

Significance. If the claims hold, the paper would be a useful advance: it provides a unified rasterization approach to several known 3DGS artifacts, with clean derivations in Appendices A and B, an open-source implementation, and a broad evaluation including out-of-distribution views. The strongest evidence is the large-FOV and multi-resolution comparisons, where the proposed method clearly outperforms MCMC and other baselines, and the timing results show that the added culling largely compensates for the cost of hierarchical sorting. The main caveats are that in-distribution metrics are not state-of-the-art, with MCMC achieving higher PSNR on Mip-NeRF 360 and Tanks & Temples, and that the claimed exactness of the frustum culling is not supported by the described implementation.

major comments (3)
  1. [Sec. 3.4, Eq. (18)] The text states that 'we compute the point of maximum contribution of the Gaussian inside this 3D frustum' and later calls the culling 'exact frustum culling', but the implementation only projects onto the two screen-space-closest x/y planes and their three associated edges, not all six faces and twelve edges. For a convex quadratic, minimizing over a boundary subset yields an upper bound on the true minimum over the full frustum, so a tile can be discarded when the true minimum is below the threshold tau_rho, meaning a Gaussian that contributes to that tile is wrongly culled. Because the screen-space-closest plane is not necessarily the closest plane in the transformed Gaussian space under the nonlinear projective map, this is not a purely theoretical concern, and it directly affects the claimed removal of popping at image borders and close-up views. The authors should either compute all planes and edges, add a conservative margin so that culling is guaranteed not to remove contributing Gaussians, or present a quantitative check that false culls never occur in the evaluated scenes.
  2. [Abstract and Sec. 4.1, Table 1] The abstract claims 'state-of-the-art reconstruction quality' and 'significantly outperforms other approaches for out-of-distribution views', but Table 1 shows that MCMC has higher PSNR on Mip-NeRF 360 (28.027 vs. 27.835) and on Tanks & Temples (24.642 vs. 23.582), and the text itself says 'our method outperforms others in nearly all metrics and matches MCMC in overall quality'. The in-distribution claim should be rephrased to match the data, e.g., 'comparable to state-of-the-art on in-distribution views while achieving better out-of-distribution robustness.'
  3. [Sec. 3.1 and Sec. 3.2, Eq. (9)-(13)] The amplitude normalization in Eq. (10) is derived under the explicit modeling choice that a pixel's Gaussian contribution is the maximum value along the viewing ray rather than an integral over the ray. This assumption is inherited from Hahlbohm et al. [11] and is stated clearly, but it is load-bearing for the new anti-aliasing filter: if a future variant of the pipeline switched to full volumetric integration, the perpendicular-area normalization would no longer be calibrated. The paper would be strengthened by stating this dependency explicitly in the limitations and by discussing whether the maximum-along-ray approximation is validated for the artifact-free claim, especially at close distances where multiple high-magnitude contributions may overlap.
minor comments (5)
  1. [Sec. 3.1] There is a typo: 'Following Kerblet al. [14]' should be 'Following Kerbl et al. [14]'.
  2. [Table 3] The Mip-Splatting row is duplicated for Mip-NeRF 360 and Tanks & Temples; one duplicate should be removed.
  3. [Sec. 3.3, Eq. (14)-(15)] The notation s1,3 and s2,3 is defined in the text, but the meaning of the subscripts is easy to confuse with the entries of the covariance matrix; using s_{i,j} = <t, T_i ⊙ T_j> consistently in the equations would improve clarity.
  4. [Sec. 4.2, 'Close to Scene Camera Location'] The paper states that popping artifacts close to scene content are difficult to evaluate quantitatively and refers to a video; a static figure with annotated frames before and after the popping event would make the claim more verifiable in the printed manuscript.
  5. [Sec. 3.4] The phrase 'reducing sorting costs' appears in the contribution list and should be 'reducing sorting costs' or 'reducing sorting overhead' for grammatical consistency.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the new filter, bounding, and culling derivations are self-contained from stated geometric and linear-algebra assumptions, and the reported results are benchmarked against external baselines rather than reducing to fitted inputs.

full rationale

I walked the paper's derivation chain and found no step in which a claimed prediction or first-principles result reduces by construction to its inputs. The adaptive 3D smoothing filter (Sec. 3.2, Eqs. 9-13, Appendix A) is derived from the determinant and Schur-complement identities under the explicit maximum-along-ray evaluation model inherited from Hahlbohm et al. [11]; no evaluation metric or test image enters the derivation. The constant k=0.3 is taken from prior work, not fitted to the reported PSNR/SSIM/LPIPS numbers. The view-space bounding (Sec. 3.3, Eqs. 14-17, Appendix B) solves the quadratic touching condition for view-space planes and is presented as closely related to Hahlbohm et al. [11] while moving the bounding before projection; this is an incremental derivation, not a renaming of the target artifact. The frustum-based culling (Sec. 3.4, Eq. 18) tests a minimum of rho(x)^2 against a threshold; the implementation restricts the search to the two closest x/y planes and three edges, so the word 'exact' is stronger than what is implemented, but that is an approximation/robustness gap, not circularity: the culling decision is not defined by the popping artifact it is meant to remove, nor is the threshold fitted to that artifact. Overlapping-author citations (StopThePop [28], Frustum Volume Caching [31], VRSplat [33]) are used as rendering bases or background, not as authority for the novel filter, bounding, or culling claims. The central contributions are therefore self-contained and externally benchmarked.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central claim rests on three new mathematical components, all derived in the appendices from standard linear algebra and projective geometry. The assumptions it pulls from prior work are the maximum-contribution ray evaluation (Hahlbohm et al.), the hierarchical sort (StopThePop), and the pinhole camera model. No new physical entities are introduced. The only hand-set constants are k=0.3 and the unstated tau_rho and epsilon values.

free parameters (3)
  • k (smoothing filter kernel size) = 0.3
    Controls the amount of 3D dilation for anti-aliasing; adopted from prior work (Kerbl et al., Yu et al.), not optimized in this paper, but it directly affects the filter behavior in Eq. 7-13.
  • tau_rho (bounding/culling threshold) = not specified
    Defines the ellipsoid level set used for plane fitting in Sec. 3.3 and for frustum culling in Sec. 3.4; no numeric value is reported, so the exact bounding tightness is underspecified.
  • epsilon (angle bound margin) = not specified
    Small margin used in Eq. 17 to clamp view-space angles to the screen range; value not reported.
assumptions (5)
  • standard math The Schur complement and block matrix inversion identities used in App. A to express the perpendicular determinant.
    Standard linear algebra, used without proof in the derivation of Eq. 10/12.
  • standard math The plane touching condition for a Gaussian ellipsoid at level set tau_rho (Sigg et al. [30]) is valid in projective space.
    Assumed as background for the view-space bounding derivation in App. B.
  • domain assumption Each pixel ray's contribution is computed as the maximum of the Gaussian along the ray, not the integral (Hahlbohm et al. [11]).
    This is the core rendering model that the new filter normalization and culling are built on; it is inherited from prior work, not re-derived.
  • domain assumption A pinhole perspective camera model with view/projection matrices V and P applies.
    The view-space bounding in Sec. 3.3 and the transformed planes rely on perspective projection; the authors acknowledge this limits generalization to other camera models.
  • domain assumption The hierarchical sort with per-pixel depth ordering from StopThePop [28] is a sufficient approximation of correct blending order.
    The pipeline builds on StopThePop's sorting and queuing, which is itself an approximation to true per-pixel sorting; the new culling must preserve its correctness.

how reviews work

0 comments
Cite this review

Pith. "Pith review of AAA-Gaussians: Anti-Aliased and Artifact-Free 3D Gaussian Rendering." pith.science (2026). https://pith.science/paper/JYPGA3IY

@misc{pith2026250412811,
  author       = {Pith},
  title        = {Pith review of: AAA-Gaussians: Anti-Aliased and Artifact-Free 3D Gaussian Rendering},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JYPGA3IY}},
  note         = {Machine review of arXiv:2504.12811}
}
read the original abstract

Although 3D Gaussian Splatting (3DGS) has revolutionized 3D reconstruction, it still faces challenges such as aliasing, projection artifacts, and view inconsistencies, primarily due to the simplification of treating splats as 2D entities. We argue that incorporating full 3D evaluation of Gaussians throughout the 3DGS pipeline can effectively address these issues while preserving rasterization efficiency. Specifically, we introduce an adaptive 3D smoothing filter to mitigate aliasing and present a stable view-space bounding method that eliminates popping artifacts when Gaussians extend beyond the view frustum. Furthermore, we promote tile-based culling to 3D with screen-space planes, accelerating rendering and reducing sorting costs for hierarchical rasterization. Our method achieves state-of-the-art quality on in-distribution evaluation sets and significantly outperforms other approaches for out-of-distribution views. Our qualitative evaluations further demonstrate the effective removal of aliasing, distortions, and popping artifacts, ensuring real-time, artifact-free rendering.

Figures

Figures reproduced from arXiv: 2504.12811 by the authors.

Figure 1
Figure 1. (Top row) 3DGS rasterization approaches encounter artifacts in out-of-distribution camera settings: (1) Distortions from 2D splat [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. (Left) Aliasing artifacts manifest when camera positions [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 5
Figure 5. (Left) Hahlbohm et al. [11] compute the screen bounds of a Gaussian by fitting planes in screen space. However, they discard Gaussians whose z-bounds (zmin,max) are outside the near/far￾planes, which can lead to popping. (Right) We instead compute view space angles θ1,2, leading to a more robust computation and bounding. Since the inverse covariance matrix is given by Σ−1 = RS−2R⊤, we can express the directional qua… view at source ↗
Figures from the paper (2 more)
Figure 7
Figure 7. Figure 7: Example results with PSNR when rendering with a larger [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: A single view of our multi-resolution evaluation on the Mip-Nerf 360 bicycle scene, with inset PSNR values. [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. A Hybrid Neural-Microfacet BRDF Model for Real-Time Rendering

    cs.GR 2026-08 conditional novelty 6.0 of 10

    A hybrid BRDF model, combining a GGX analytical term with a tiny learned residual and gating network, fits measured materials more accurately than fully neural models at equal memory cost.

Reference graph

Works this paper leans on

41 extracted references · 38 canonical work pages · cited by 1 Pith paper

  1. [11]

    Efficient Perspective- Correct 3D Gaussian Splatting Using Hybrid Transparency,

    Florian Hahlbohm, Fabian Friederichs, Tim Weyrich, Linus Franke, Moritz Kappel, Susana Castillo, Marc Stamminger, Martin Eisemann, and Marcus Magnor. Efficient Perspective- Correct 3D Gaussian Splatting Using Hybrid Transparency,

    work page

  2. [1]

    Barron, Ben Mildenhall, Matthew Tancik, Peter Hedman, Ricardo Martin-Brualla, and Pratul P

    Jonathan T. Barron, Ben Mildenhall, Matthew Tancik, Peter Hedman, Ricardo Martin-Brualla, and Pratul P. Srinivasan. Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fields. In IEEE/CVF International Confer- ence on Computer Vision, 2021. 2

    work page 2021

  3. [2]

    Barron, Ben Mildenhall, Dor Verbin, Pratul P

    Jonathan T. Barron, Ben Mildenhall, Dor Verbin, Pratul P. Srinivasan, and Peter Hedman. Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields. In IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition, 2022. 2, 6

    work page 2022

  4. [3]

    Barron, Ben Mildenhall, Dor Verbin, Pratul P

    Jonathan T. Barron, Ben Mildenhall, Dor Verbin, Pratul P. Srinivasan, and Peter Hedman. Zip-NeRF: Anti-Aliased Grid- Based Neural Radiance Fields. In IEEE/CVF International Conference on Computer Vision, 2023. 2

    work page 2023

  5. [4]

    RaySplats: Ray Tracing based Gaussian Splatting

    Krzysztof Byrski, Marcin Mazur, Jacek Tabor, Tadeusz Dziar- maga, Marcin K ˛ adziołka, Dawid Baran, and Przemysław Spurek. RaySplats: Ray Tracing based Gaussian Splatting. arXiv preprint arXiv:2501.19196, 2025. 2

    arXiv 2025

  6. [5]

    Does 3D Gaussian Splatting Need Accurate V olumetric Rendering? InComputer Graphics Forum, page e70032, 2025

    Adam Celarek, George Kopanas, George Drettakis, Michael Wimmer, and Bernhard Kerbl. Does 3D Gaussian Splatting Need Accurate V olumetric Rendering? InComputer Graphics Forum, page e70032, 2025. 2

    work page 2025

  7. [6]

    Don’t Splat your Gaussians: V olumetric Ray-Traced Primitives for Mod- eling and Rendering Scattering and Emissive Media

    Jorge Condor, Sebastien Speierer, Lukas Bode, Aljaz Bozic, Simon Green, Piotr Didyk, and Adrian Jarabo. Don’t Splat your Gaussians: V olumetric Ray-Traced Primitives for Mod- eling and Rendering Scattering and Emissive Media. ACM TOG, 2025. 2

    work page 2025

  8. [7]

    Daniel Duckworth, Peter Hedman, Christian Reiser, Pe- ter Zhizhin, Jean-François Thibert, Mario Lu ˇci´c, Richard Szeliski, and Jonathan T. Barron. SMERF: Streamable Mem- ory Efficient Radiance Fields for Real-Time Large-Scene Exploration. ACM TOG, 43(4), 2024. 2

    work page 2024

Show all 41 references
  1. [8]

    LightGaussian: Unbounded 3D Gaussian Compression with 15x Reduction and 200+ FPS

    Zhiwen Fan, Kevin Wang, Kairun Wen, Zehao Zhu, Dejia Xu, and Zhangyang Wang. LightGaussian: Unbounded 3D Gaussian Compression with 15x Reduction and 200+ FPS. In Advances in Neural Information Processing Systems, 2024. 2

    2024

  2. [9]

    Mini-Splatting: Representing Scenes with a Constrained Number of Gaussians

    Guangchi Fang and Bing Wang. Mini-Splatting: Representing Scenes with a Constrained Number of Gaussians. InEuropean Conference on Computer Vision, 2024. 2

    2024

  3. [10]

    Plenoxels: Radiance Fields without Neural Networks

    Sara Fridovich-Keil, Alex Yu, Matthew Tancik, Qinhong Chen, Benjamin Recht, and Angjoo Kanazawa. Plenoxels: Radiance Fields without Neural Networks. In IEEE/CVF Conference on Computer Vision and Pattern Recognition ,

  4. [12]

    Deep Blending for Free-viewpoint Image-based Rendering

    Peter Hedman, Julien Philip, True Price, Jan-Michael Frahm, George Drettakis, and Gabriel Brostow. Deep Blending for Free-viewpoint Image-based Rendering. ACM TOG, 37(6),

  5. [13]

    On the Error Analysis of 3D Gaussian Splatting and an Optimal Projection Strategy

    Letian Huang, Jiayang Bai, Jie Guo, Yuanqi Li, and Yanwen Guo. On the Error Analysis of 3D Gaussian Splatting and an Optimal Projection Strategy. In European Conference on Computer Vision, 2024. 2, 3, 7

    2024

  6. [14]

    3D Gaussian Splatting for Real-Time Radi- ance Field Rendering

    Bernhard Kerbl, Georgios Kopanas, Thomas Leimkühler, and George Drettakis. 3D Gaussian Splatting for Real-Time Radi- ance Field Rendering. ACM TOG, 42(4), 2023. 1, 2, 3, 4, 6, 7

    2023

  7. [15]

    A Hierarchical 3D Gaussian Representation for Real-Time Rendering of Very Large Datasets

    Bernhard Kerbl, Andreas Meuleman, Georgios Kopanas, Michael Wimmer, Alexandre Lanvin, and George Drettakis. A Hierarchical 3D Gaussian Representation for Real-Time Rendering of Very Large Datasets. ACM TOG, 43(4), 2024. 2

    2024

  8. [16]

    3D Gaussian Splatting as Markov Chain Monte Carlo

    Shakiba Kheradmand, Daniel Rebain, Gopal Sharma, Weiwei Sun, Yang-Che Tseng, Hossam Isack, Abhishek Kar, Andrea Tagliasacchi, and Kwang Moo Yi. 3D Gaussian Splatting as Markov Chain Monte Carlo. In Advances in Neural Informa- tion Processing Systems, 2024. 2, 6, 7

    2024

  9. [17]

    Tanks and Temples: Benchmarking Large-Scale Scene Reconstruction

    Arno Knapitsch, Jaesik Park, Qian-Yi Zhou, and Vladlen Koltun. Tanks and Temples: Benchmarking Large-Scale Scene Reconstruction. ACM TOG, 36(4), 2017. 6

    2017

  10. [18]

    Analytic-Splatting: Anti-Aliased 3D Gaus- sian Splatting via Analytic Integration

    Zhihao Liang, Qi Zhang, Wenbo Hu, Ying Feng, Lei Zhu, and Kui Jia. Analytic-Splatting: Anti-Aliased 3D Gaus- sian Splatting via Analytic Integration. arXiv preprint arXiv:2403.11056, 2024. 2

    2024 arXiv

  11. [19]

    Barron, and Yinda Zhang

    Alexander Mai, Peter Hedman, George Kopanas, Dor Verbin, David Futschik, Qiangeng Xu, Falko Kuester, Jonathan T. Barron, and Yinda Zhang. EVER: Exact V olumetric Ellipsoid Rendering for Real-time View Synthesis, 2024. 2

    2024

  12. [20]

    Taming 3DGS: High-Quality Radiance Fields with Limited Resources

    Saswat Mallick, Rahul Goel, Bernhard Kerbl, Francisco Vicente Carrasco, Markus Steinberger, and Fernando De La Torre. Taming 3DGS: High-Quality Radiance Fields with Limited Resources. In SIGGRAPH Asia Conference Papers, 2024. 2, 6, 7

    2024

  13. [21]

    Hybrid transparency

    Marilena Maule, João Comba, Rafael Torchelsen, and Rui Bastos. Hybrid transparency. InACM SIGGRAPH Symposium on Interactive 3D Graphics and Games, pages 103–118, 2013. 2

    2013

  14. [22]

    Srinivasan, Matthew Tancik, Jonathan T

    Ben Mildenhall, Pratul P. Srinivasan, Matthew Tancik, Jonathan T. Barron, Ravi Ramamoorthi, and Ren Ng. NeRF: Representing Scenes as Neural Radiance Fields for View Syn- thesis. In European Conference on Computer Vision, 2020. 2

    2020

  15. [23]

    3D Gaussian Ray Tracing: Fast Tracing of Particle Scenes

    Nicolas Moenne-Loccoz, Ashkan Mirzaei, Or Perel, Riccardo de Lutio, Janick Martinez Esturo, Gavriel State, Sanja Fidler, Nicholas Sharp, and Zan Gojcic. 3D Gaussian Ray Tracing: Fast Tracing of Particle Scenes. ACM TOG, 2024. 1

    2024

  16. [24]

    3D Gaussian Ray Tracing: Fast Tracing of Particle Scenes

    Nicolas Moenne-Loccoz, Ashkan Mirzaei, Riccardo Perel, or de Lutio, Janick Martinez Esturo, Gavriel State, Sanja Fidler, Nicholas Sharo, and Zan Gojcic. 3D Gaussian Ray Tracing: Fast Tracing of Particle Scenes. ACM TOG, 43(6), 2024. 2

    2024

  17. [25]

    Instant Neural Graphics Primitives with a Mul- tiresolution Hash Encoding

    Thomas Müller, Alex Evans, Christoph Schied, and Alexan- der Keller. Instant Neural Graphics Primitives with a Mul- tiresolution Hash Encoding. ACM TOG, 41(4), 2022. 2

    2022

  18. [26]

    Mueller, Chakravarty R

    Thomas Neff, Pascal Stadlbauer, Mathias Parger, Andreas Kurz, Joerg H. Mueller, Chakravarty R. Alla Chaitanya, An- ton Kaplanyan, and Markus Steinberger. DONeRF: Towards Real-Time Rendering of Compact Neural Radiance Fields using Depth Oracle Networks. In Computer Graphics For...

    2021

  19. [27]

    RadSplat: Radiance Field-Informed Gaussian Splat- ting for Robust Real-Time Rendering with 900+ FPS

    Michael Niemeyer, Fabian Manhardt, Marie-Julie Rakoto- saona, Michael Oechsle, Daniel Duckworth, Rama Gosula, Keisuke Tateno, John Bates, Dominik Kaeser, and Federico Tombari. RadSplat: Radiance Field-Informed Gaussian Splat- ting for Robust Real-Time Rendering with 900+ FPS. ...

    2024 arXiv

  20. [28]

    StopThePop: Sorted Gaussian Splatting for View-Consistent Real-time Rendering

    Lukas Radl, Michael Steiner, Mathias Parger, Alexan- der Weinrauch, Bernhard Kerbl, and Markus Steinberger. StopThePop: Sorted Gaussian Splatting for View-Consistent Real-time Rendering. ACM TOG, 43(4), 2024. 1, 2, 3, 4, 6, 7

    2024

  21. [29]

    Schönberger and Jan-Michael Frahm

    Johannes L. Schönberger and Jan-Michael Frahm. Structure- from-Motion Revisited. In IEEE/CVF Conference on Com- puter Vision and Pattern Recognition, 2016. 2

    2016

  22. [30]

    GPU-Based Ray-Casting of Quadratic Surfaces

    Christian Sigg, Tim Weyrich, Mario Botsch, and Markus Gross. GPU-Based Ray-Casting of Quadratic Surfaces. In Symposium on Point-Based Graphics. The Eurographics As- sociation, 2006. 3, 5, 11

    2006

  23. [31]

    Frustum V olume Caching for Accelerated NeRF Rendering

    Michael Steiner, Thomas Köhler, Lukas Radl, and Markus Steinberger. Frustum V olume Caching for Accelerated NeRF Rendering. Proceedings of the ACM on Computer Graphics and Interactive Technologies, 7(3), 2024. 2

    2024

  24. [32]

    V olumetrically Consistent 3D Gaussian Rasterization, 2025

    Chinmay Talegaonkar, Yash Belhe, Ravi Ramamoorthi, and Nicholas Antipa. V olumetrically Consistent 3D Gaussian Rasterization, 2025. 1, 3

    2025

  25. [33]

    VRSplat: Fast and Robust Gaussian Splatting for Virtual Reality

    Xuechang Tu, Lukas Radl, Michael Steiner, Markus Stein- berger, Bernhard Kerbl, and Fernando de la Torre. VRSplat: Fast and Robust Gaussian Splatting for Virtual Reality. Pro- ceedings of the ACM on Computer Graphics and Interactive Technologies, 8(1), 2025. 2

    2025

  26. [34]

    3DGUT: Enabling Dis- torted Cameras and Secondary Rays in Gaussian Splatting,

    Qi Wu, Janick Martinez Esturo, Ashkan Mirzaei, Nicolas Moenne-Loccoz, and Zan Gojcic. 3DGUT: Enabling Dis- torted Cameras and Secondary Rays in Gaussian Splatting,

  27. [35]

    Multi-scale 3d gaussian splatting for anti-aliased rendering

    Zhiwen Yan, Weng Fei Low, Yu Chen, and Gim Hee Lee. Multi-scale 3d gaussian splatting for anti-aliased rendering. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 20923–20931, 2024. 2

    2024

  28. [36]

    Mip-Splatting: Alias-free 3D Gaussian Splat- ting

    Zehao Yu, Anpei Chen, Binbin Huang, Torsten Sattler, and Andreas Geiger. Mip-Splatting: Alias-free 3D Gaussian Splat- ting. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024. 2, 3, 4, 5, 6, 7

    2024

  29. [37]

    Gaussian Opacity Fields: Efficient Adaptive Surface Reconstruction in Unbounded Scenes

    Zehao Yu, Torsten Sattler, and Andreas Geiger. Gaussian Opacity Fields: Efficient Adaptive Surface Reconstruction in Unbounded Scenes. ACM TOG, 2024. 1, 2, 3, 4

    2024

  30. [38]

    The Unreasonable Effectiveness of Deep Features as a Perceptual Metric

    Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shechtman, and Oliver Wang. The Unreasonable Effectiveness of Deep Features as a Perceptual Metric. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018. 6

    2018

  31. [39]

    EW A V olume Splatting

    Mathias Zwicker, Hanspeter Pfister, Jeroen van Baar, and Markus Gross. EW A V olume Splatting. InIEEE Visualization,

  32. [2001]

    Derivation of Amplitude Scaling Factor Let d = µ− o ∥µ− o∥ be a unit vector in R3, where µ is the mean of the Gaussian and o is the camera position in world space

    2, 3, 6 A. Derivation of Amplitude Scaling Factor Let d = µ− o ∥µ− o∥ be a unit vector in R3, where µ is the mean of the Gaussian and o is the camera position in world space. We are inter- ested in the area of the Gaussian’s intersection with the plane perpendicular to d. Let ...

  33. [2024]

    1, 2, 3, 4, 5, 6, 7, 11

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.