Pith. sign in

REVIEW 3 major objections 5 minor 2 cited by

AD-GS: Object-Aware B-Spline Gaussian Splatting for Self-Supervised Autonomous Driving

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read AD-GS claims that an annotation-free model can render dynamic driving scenes as well as tracklet-supervised methods, reporting 29.16 PSNR on KITTI-75% and 33.91 PSNR on Waymo with a B-spline-plus-trigonometric motion model.

desk verdict A solid self-supervised driving-scene renderer with a genuinely novel motion basis, but the reported gains rest on a one-time pseudo-mask split that the paper leaves untested. read the letter →

arxiv 2507.12137 v3 pith:ABWUOOBS submitted 2025-07-16 cs.CV

Jiawei Xu , Kai Deng , Zexin Fan , Shenlong Wang , Jin Xie , Jian Yang This is my paper
classification cs.CV
keywords autonomousdrivingscenerendering3DGaussiansplattingself-supervisedlearningB-splinecurvesdynamicdecompositiontemporalvisibilityfree-viewpointLiDAR-guidedsupervision
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

AD-GS is a self-supervised system for free-viewpoint rendering of urban driving logs. It claims that manual 3D object tracklets are unnecessary: a two-class split of the scene into moving objects and static background, driven by text-prompted pseudo segmentation, is enough to reach rendering quality that is competitive with supervised methods. The key move is replacing pure trigonometric or network motion models with learnable B-spline curves, plus quaternion B-splines for rotation, combined with global trigonometric terms, so that each Gaussian's trajectory can fit both global motion and local detail. On KITTI-75% it reports PSNR 29.16 versus 27.13 for the best self-supervised baseline, and on Waymo it lands within about 0.7 dB of the strongest annotation-dependent method. If correct, this would remove a major annotation bottleneck for driving simulation and closed-loop testing.

What carries the argument

The load-bearing object is the learnable motion model for each dynamic Gaussian: a uniform B-spline curve for position, defined by learnable control points and evaluated through a precomputed matrix formulation, together with a B-spline quaternion curve for rotation, both added to trigonometric terms. A B-spline is a piecewise polynomial curve whose value at any time depends only on nearby control points, so optimizing a frame's motion updates only a few control points and gives local fitting, while the trigonometric terms provide global fitting that resists noise from pseudo labels. The second mechanism is the bidirectional temporal visibility mask, which multiplies an object Gaussian's opacity by a Gaussian in time centered at the LiDAR acquisition timestamp, with separate learnable widths for before and after, so that suddenly appearing or disappearing objects do not receive misleading gradients from frames where they are invisible.

What would settle it

A decisive check is to run the released model on a KITTI sequence after changing the object text prompt to a class that excludes the actual moving vehicles; if the pipeline still delivers near-29 PSNR on KITTI-75%, the object/background decomposition is not the source of the reported gains, whereas a sharp drop would confirm that the simplified pseudo segmentation is load-bearing.

Watch

Extended reading notes

Core claim

The paper's central discovery is that the two failure modes of self-supervised driving-scene rendering, noisy scene decomposition and inaccurate local motion, can be addressed together without any manual 3D supervision. AD-GS partitions all Gaussians into objects and background using a simplified pseudo 2D segmentation, keeps the background stationary, and deforms object Gaussians over time using learnable B-spline curves combined with trigonometric functions. Each object Gaussian also carries a bidirectional temporal visibility mask centered at the LiDAR acquisition timestamp, which lets the model ignore frames where the object is not actually visible. Combined with flow, inverse-depth, visibility, and rigidity supervision, the model reports state-of-the-art self-supervised results and competitive results against annotation-based systems on KITTI, Waymo, and nuScenes.

Load-bearing premise

The load-bearing premise is that the simplified pseudo 2D segmentation and the LiDAR acquisition timestamp correctly identify which Gaussians are dynamic and when each object is visible, so that the motion model and temporal visibility mask receive a correct separation of moving and static content.

Editorial extensions

If this is right

  • A single driving log can be turned into a free-viewpoint, free-time rendering without any manual 3D tracklets, at speeds around 47 FPS on KITTI.
  • Two-class decomposition into objects and background is enough to isolate motion while avoiding the noise of fine-grained instance segmentation.
  • The combined B-spline and trigonometric motion model captures both global trajectory and local detail, improving PSNR over trigonometric-only rendering by about 2 dB on KITTI-75%.
  • Bidirectional temporal visibility masks reduce artifacts when objects appear or disappear, preventing invisible frames from corrupting the motion fit.
  • Even with only 25% of KITTI frames for training, the method still improves over self-supervised baselines, suggesting resilience to sparse supervision.

Reading between the lines

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

  • A testable extension beyond the paper: the same two-class, text-prompted decomposition should transfer to other dynamic scene logs with small known sets of moving classes, but the paper only demonstrates urban driving.
  • The use of the LiDAR acquisition timestamp as a fixed prompt for visibility turns sensor timing into a free supervisory signal; one could extend it to cameras with rolling shutters, where visibility time varies per row.
  • Because the loss ablation shows segmentation guidance is the largest single lever, improving open-vocabulary segmentation quality should push the same architecture closer to or past annotation-dependent models without any change to the motion model.
  • The local-fitting argument implies a specific prediction: on objects visible for only a handful of frames, AD-GS should degrade more gracefully than trigonometric-only models, which can be tested directly on very short-lived objects.
Share X Bluesky LinkedIn Reddit HN

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 proposes AD-GS, a Gaussian-splatting framework for dynamic urban scene rendering without manual 3D annotations. LiDAR-initialized Gaussians are partitioned once into object and background components using Grounded-SAM binary masks; object Gaussians are deformed by a combination of B-spline curves, B-spline quaternion curves, and trigonometric functions, and are gated by bidirectional temporal visibility masks; background Gaussians remain geometrically static. Training is self-supervised with image reconstruction, rendered object-mask supervision, optical-flow supervision, inverse-depth supervision, and regularization losses. Experiments on KITTI, Waymo, and nuScenes compare AD-GS with self-supervised baselines (SUDS, EmerNeRF, PVG, Grid4D) and annotation-assisted methods, reporting consistent improvements over the self-supervised baselines and near-parity with supervised StreetGS/4DGF on Waymo.

Significance. If the results hold, AD-GS is a meaningful step toward annotation-free dynamic driving-scene rendering: on KITTI-75% it reports PSNR 29.16 / SSIM 0.920 / LPIPS 0.033 versus PVG's 27.13 / 0.895 / 0.049, and on Waymo it reaches PSNR 33.91, close to StreetGS's 33.97, while remaining real-time. The paper is strong in evaluating on three public benchmarks with held-out frames, including ablations of each loss term and module, and it reports rendering speed. However, the central decomposition into object and background relies on a one-shot pseudo-mask partition whose robustness is not quantified, baseline hyperparameters are not fully controlled on nuScenes, and no error bars or per-sequence results are provided. These points do not invalidate the method but limit the strength of the current claims.

major comments (3)
  1. [Sec. 3.3 and Sec. A (Supplementary)] The object/background decomposition is decided once at initialization from Grounded-SAM binary masks and is never revised: the attribute-inheritance rule in Sec. A ensures that densified Gaussians keep their parent's label, and Lobj in Eq. (9) only adjusts opacity. Consequently, any dynamic object missed by the hand-written prompt list ('car.bus.truck.van.human', plus 'bike' on nuScenes) or by the mask/projection step is permanently static and is never deformed by Eq. (7). This is load-bearing because the motion model applies only to object Gaussians and the flow supervision in Sec. 3.4 is restricted to the object mask. The paper's own Sec. D acknowledges that low pseudo-label quality produces artifacts (Fig. 10), but no experiment measures how sensitive the reported margins over PVG/EmerNeRF are to this fixed split. Please add a quantitative sensitivity analysis, e.g., train with ground-truth object masks as an oracle, drop individual prompt classes, or corrupt/mask part of the segmentation, and report the resulting PSNR/SSIM/LPIPS on KITTI or Waymo.
  2. [Supplementary Sec. B (Baseline details)] PVG is evaluated on nuScenes with hyperparameters designed for Waymo, and EmerNeRF's hyperparameters are not described as tuned for nuScenes. Dynamic-scene Gaussian methods are sensitive to sampling densities, deformation network capacity, and opacity thresholds, so the nuScenes comparison in Table 3 is not fully controlled. Since the abstract claims significant improvement over state-of-the-art annotation-free methods, please retune both baselines on nuScenes, or supply a sensitivity analysis showing that the reported numbers are robust to hyperparameter choices, and report the settings used for every dataset.
  3. [Tables 1-3] The main tables report averages over sequences (3 KITTI logs, 8 Waymo logs, 6 nuScenes logs) without standard deviations or per-sequence numbers. This makes it impossible to judge whether the reported margins, e.g., PSNR 29.16 vs 27.13 on KITTI-75%, are consistent across logs or dominated by a single sequence. Please report per-sequence results and standard deviations for the main comparison tables so that the word 'significantly' in the abstract is supported statistically.
minor comments (5)
  1. [Sec. 3.3, Eq. (9)] The sentence 'Lobj makes the two Gaussian subsets stayed in their respective areas' is grammatically unclear; please rephrase to say that the loss encourages the subsets to remain in their assigned regions.
  2. [Abstract and Sec. 3.3] The abstract says the scene is 'automatically segmented', but the segmentation is driven by hand-written text prompts listed in Sec. A; this dependence should be acknowledged in the main text.
  3. [Conclusion / Supp. Sec. D] The limitations section is relegated to the supplementary material; given its importance for the one-shot segmentation assumption, consider moving a condensed version into the main paper.
  4. [Sec. 3.1 and Supp. Sec. A] The paper states background Gaussians 'remain stationary' while Eq. (3) deforms color features of all Gaussians; clarify that 'stationary' refers to position and rotation, not appearance.
  5. [General] No code is released; please include a statement about code availability or provide a reproducibility appendix with the exact training schedules and per-sequence settings.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: AD-GS's rendering results are evaluated on held-out frames of public benchmarks, and the only self-citation (Grid4D) appears as a baseline and visualization convention, not as load-bearing evidence.

full rationale

The paper's central performance claim is external rather than circular: AD-GS is compared with baselines on held-out test frames of KITTI, Waymo, and nuScenes (Tables 1-3), so the reported PSNR/SSIM/LPIPS values are not produced by the training objective or by the fitted parameters being fed back as the answer. The B-spline-plus-trigonometric motion model (Eq. 7) and the temporal visibility mask (Eq. 10) are optimized on training frames and then evaluated at test timestamps; no equation in the manuscript reduces the test-set metric to a training loss. The object/background partition is a one-time initialization from external Grounded-SAM pseudo masks (Sec. 3.3 and Sec. A), and Lobj is a training loss that aligns the rendered mask to those same pseudo labels; this is a pseudo-label-supervised design choice, not a self-referential derivation, because the claimed novelty is rendering quality, not segmentation prediction. The paper's Section D limitations (artifacts under low pseudo-label quality, failures on briefly visible objects) and the Attribute Inheritance paragraph in Sec. A describe robustness boundaries of this choice, but they do not make the rendering result equivalent to its inputs. The only self-citation is Grid4D [34], which shares authors; it appears as a baseline ('We also include Grid4D [34], a state-of-the-art model for object-level dynamic scene rendering, as a baseline') and as a visualization convention ('we visualize the deformation map mainly following the approach in Grid4D [34]'). This citation is not load-bearing: removing it would not change the comparisons against PVG, EmerNeRF, StreetGS, or 4DGF, and it is not used to justify the motion model or the decomposition. No uniqueness theorem is imported, no fitted parameter is renamed as a prediction, and no ansatz is smuggled in via a self-citation. The score is therefore 2, reflecting only a minor non-load-bearing self-citation.

Assumptions & free parameters 5 free parameters · 5 assumptions · 2 invented entities

The central claim rests on a standard Gaussian splatting forward model, a standard B-spline/quaternion construction, and strong domain assumptions about pseudo-label quality, the LiDAR timestamp as visibility prompt, and local rigidity. Five hyperparameter choices (B-spline order, control point density, trigonometric frequency, loss weights, segmentation prompts) are hand-set or ablation-selected and then fixed across datasets, which limits the generality of the reported margins but is not itself circular.

free parameters (5)
  • B-spline order k = 6
    Chosen by ablation on KITTI (Table 7); all datasets use k=6; higher or lower orders give similar metrics, so not a strong lever.
  • Control point count (one per 3 frames) = n+1 = total_frames/3
    Selected by ablation (Table 6); too many control points overfit noisy pseudo labels, too few lose local detail.
  • Trigonometric frequency K, L = 6
    Fixed hand-set value for color and position deformation (Eq. 3 and Eq. 7); no ablation reported.
  • Loss weights lambda_c, lambda_d, lambda_f, lambda_obj, lambda_sky, lambda_r, lambda_s = 0.2, 0.1, 0.1, 0.1, 0.05, 0.5, 0.01
    Hand-set across all experiments (Eq. 16); no sensitivity analysis reported.
  • Object class prompts for segmentation = 'car.bus.truck.van.human' (KITTI/Waymo) plus 'bike' (nuScenes)
    Manually chosen text prompts for Grounded-SAM (Supplementary A); determines which Gaussians are treated as dynamic objects.
assumptions (5)
  • standard math The 3D Gaussian splatting forward model (Eq. 1 and Eq. 2) and its differentiable rasterizer are correct and applicable to driving scenes.
    Taken from Kerbl et al. [9]; the paper relies on it without derivation (Sec 3.1).
  • standard math Uniform B-spline matrix form (Eq. 5, 6) and quaternion B-spline construction (Eq. 8) are valid for modeling position and rotation deformations.
    From Qin [25] and Kim et al. [11]; no proof in the paper, used as building blocks (Sec 3.2).
  • domain assumption Pseudo-labels from CoTracker3, DPTv2, and Grounded-SAM are sufficiently accurate to supervise motion, depth, and segmentation.
    Sec 3.4 and Supplementary A; the paper's Limitations section D admits artifacts when pseudo-label quality is low.
  • ad hoc to paper The LiDAR acquisition timestamp mu_t of each Gaussian is a correct prompt for when the object is visible.
    Eq. 10; the model treats mu_t as fixed and learnable s0, s1 control the visibility interval; this is the paper's own modeling choice, not derived.
  • domain assumption Local rigidity: nearby object Gaussians should have similar deformation and visibility parameters.
    Sec 3.4 Eq. 15 KNN variance; the assumption is not physically enforced as a hard rigid transform.
invented entities (2)
  • Bidirectional temporal visibility mask on Gaussian opacity
    purpose: Suppress a dynamic Gaussian when its object is not visible (before or after appearance), to avoid wrong motion gradients from invisible frames.
    Internal model component with no externally falsifiable prediction; its effect is judged only through rendering metrics (Eq. 10).
  • Simplified two-class object/background Gaussian decomposition driven by text prompts
    purpose: Partition scene Gaussians into dynamic objects and static background without manual 3D boxes.
    A representational construct, not a physical entity; the object class list is hand-selected (Supplementary A).

how reviews work

0 comments
Cite this review

Pith. "Pith review of AD-GS: Object-Aware B-Spline Gaussian Splatting for Self-Supervised Autonomous Driving." pith.science (2026). https://pith.science/paper/ABWUOOBS

@misc{pith2026250712137,
  author       = {Pith},
  title        = {Pith review of: AD-GS: Object-Aware B-Spline Gaussian Splatting for Self-Supervised Autonomous Driving},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ABWUOOBS}},
  note         = {Machine review of arXiv:2507.12137}
}
read the original abstract

Modeling and rendering dynamic urban driving scenes is crucial for self-driving simulation. Current high-quality methods typically rely on costly manual object tracklet annotations, while self-supervised approaches fail to capture dynamic object motions accurately and decompose scenes properly, resulting in rendering artifacts. We introduce AD-GS, a novel self-supervised framework for high-quality free-viewpoint rendering of driving scenes from a single log. At its core is a novel learnable motion model that integrates locality-aware B-spline curves with global-aware trigonometric functions, enabling flexible yet precise dynamic object modeling. Rather than requiring comprehensive semantic labeling, AD-GS automatically segments scenes into objects and background with the simplified pseudo 2D segmentation, representing objects using dynamic Gaussians and bidirectional temporal visibility masks. Further, our model incorporates visibility reasoning and physically rigid regularization to enhance robustness. Extensive evaluations demonstrate that our annotation-free model significantly outperforms current state-of-the-art annotation-free methods and is competitive with annotation-dependent approaches.

Figures

Figures reproduced from arXiv: 2507.12137 by the authors.

Figure 1
Figure 1. AD-GS. We achieve high-quality rendering by self-supervised manners for autonomous driving scenes [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Comparison between traditional methods and the B [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. The pipeline of AD-GS. We decompose the scene into two components: objects and background, and initialize the Gaussians [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (10 more)
Figure 4
Figure 4. Figure 4: Qualitative comparisons on the KITTI [7] dataset. EmerNeRF Grid4D AD-GS(Ours) Ground Truth [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Qualitative comparisons on the Waymo [29] dataset. EmerNeRF Grid4D AD-GS(Ours) Ground Truth [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Qualitative comparisons on the nuScenes [ [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Visualization of loss ablation on the KITTI [ [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: Visualization of object modeling module ablation on the [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 10
Figure 10. Figure 10: Failure cases when facing complex objects (a) and ob [PITH_FULL_IMAGE:figures/full_fig_p012_10.png]
Figure 9
Figure 9. Figure 9: Deformation map of the object modeling module abla [PITH_FULL_IMAGE:figures/full_fig_p012_9.png]
Figure 12
Figure 12. Figure 12: Additional qualitative comparisons on the Waymo [ [PITH_FULL_IMAGE:figures/full_fig_p014_12.png]
Figure 13
Figure 13. Figure 13: Additional qualitative comparisons on the KITTI [ [PITH_FULL_IMAGE:figures/full_fig_p015_13.png]
Figure 14
Figure 14. Figure 14: Additional qualitative comparisons on the nuScenes [ [PITH_FULL_IMAGE:figures/full_fig_p015_14.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

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

  1. From Articulated Kinematics to Routed Visual Control for Action-Conditioned Surgical Video Generation

    cs.CV 2026-05 unverdicted novelty 7.0 of 10

    A kinematic-to-visual lifting paradigm combined with hierarchically routed control generates action-conditioned surgical videos with better faithfulness, fidelity, and efficiency.

  2. OmniNWM: Omniscient Driving Navigation World Models

    cs.CV 2025-10 conditional novelty 6.0 of 10

    OmniNWM jointly generates long panoramic multi-modal driving videos, controls them precisely via normalized Plücker ray-maps, and derives dense driving rewards from generated 3D occupancy.

Reference graph

Works this paper leans on

45 extracted references · 27 canonical work pages · cited by 2 Pith papers

  1. [1]

    nuScenes: A multi- modal dataset for autonomous driving

    Holger Caesar, Varun Bankiti, Alex H Lang, Sourabh V ora, Venice Erin Liong, Qiang Xu, Anush Krishnan, Yu Pan, Gi- ancarlo Baldan, and Oscar Beijbom. nuScenes: A multi- modal dataset for autonomous driving. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11621–11631, 2020. 6, 7, 8, 1, 3, 5

    work page 2020

  2. [2]

    Periodic Vibration Gaussian: Dynamic urban scene reconstruction and real-time rendering

    Yurui Chen, Chun Gu, Junzhe Jiang, Xiatian Zhu, and Li Zhang. Periodic Vibration Gaussian: Dynamic urban scene reconstruction and real-time rendering. arXiv preprint arXiv:2311.18561, 2023. 2, 3, 4, 6, 7, 8, 1

    arXiv 2023

  3. [3]

    OmniRe: Omni urban scene reconstruction

    Ziyu Chen, Jiawei Yang, Jiahui Huang, Riccardo de Lutio, Janick Martinez Esturo, Boris Ivanovic, Or Litany, Zan Go- jcic, Sanja Fidler, Marco Pavone, Li Song, and Yue Wang. OmniRe: Omni urban scene reconstruction. arXiv preprint arXiv:2408.16760, 2024. 1, 2

    arXiv 2024

  4. [4]

    Masked-attention mask transformer for universal image segmentation

    Bowen Cheng, Ishan Misra, Alexander G Schwing, Alexan- der Kirillov, and Rohit Girdhar. Masked-attention mask transformer for universal image segmentation. In Proceed- ings of the IEEE/CVF conference on computer vision and pattern recognition, pages 1290–1299, 2022. 3

    work page 2022

  5. [5]

    Dy- namic 3D Gaussian fields for urban areas

    Tobias Fischer, Jonas Kulhanek, Samuel Rota Bul`o, Lorenzo Porzi, Marc Pollefeys, and Peter Kontschieder. Dy- namic 3D Gaussian fields for urban areas. arXiv preprint arXiv:2406.03175, 2024. 1, 2, 6, 7, 8

    arXiv 2024

  6. [6]

    Multi-level neural scene graphs for dynamic urban environments

    Tobias Fischer, Lorenzo Porzi, Samuel Rota Bulo, Marc Pollefeys, and Peter Kontschieder. Multi-level neural scene graphs for dynamic urban environments. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 21125–21135, 2024. 1, 2, 6, 7

    work page 2024

  7. [7]

    Vision meets robotics: The KITTI dataset

    Andreas Geiger, Philip Lenz, Christoph Stiller, and Raquel Urtasun. Vision meets robotics: The KITTI dataset. The International Journal of Robotics Research , 32(11):1231– 1237, 2013. 6, 7, 8, 1, 2, 3, 5

    work page 2013

  8. [8]

    Co- Tracker3: Simpler and better point tracking by pseudo- labelling real videos

    Nikita Karaev, Iurii Makarov, Jianyuan Wang, Natalia Neverova, Andrea Vedaldi, and Christian Rupprecht. Co- Tracker3: Simpler and better point tracking by pseudo- labelling real videos. arXiv preprint arXiv:2410.11831 ,

    arXiv

Show all 45 references
  1. [9]

    3D Gaussian splatting for real-time radiance field rendering

    Bernhard Kerbl, Georgios Kopanas, Thomas Leimk ¨uhler, and George Drettakis. 3D Gaussian splatting for real-time radiance field rendering. ACM Trans. Graph., 42(4):139–1,

  2. [10]

    A hierarchical 3D Gaussian representation for real-time ren- dering of very large datasets.ACM Transactions on Graphics (TOG), 43(4):1–15, 2024

    Bernhard Kerbl, Andreas Meuleman, Georgios Kopanas, Michael Wimmer, Alexandre Lanvin, and George Drettakis. A hierarchical 3D Gaussian representation for real-time ren- dering of very large datasets.ACM Transactions on Graphics (TOG), 43(4):1–15, 2024. 5

    2024

  3. [11]

    A general construction scheme for unit quaternion curves with simple high order derivatives

    Myoung-Jun Kim, Myung-Soo Kim, and Sung Yong Shin. A general construction scheme for unit quaternion curves with simple high order derivatives. InProceedings of the 22nd an- nual conference on Computer graphics and interactive tech- niques, pages 369–376, 1995. 4

    1995

  4. [12]

    Segment any- thing

    Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer White- head, Alexander C Berg, Wan-Yen Lo, et al. Segment any- thing. In Proceedings of the IEEE/CVF International Con- ference on Computer Vision, pages 4015–4026, 202...

    2023

  5. [13]

    Imagenet classification with deep convolutional neural net- works

    Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. Imagenet classification with deep convolutional neural net- works. Advances in neural information processing systems , 25, 2012. 6

    2012

  6. [14]

    Panoptic Neural Fields: A semantic object-aware neural scene representation

    Abhijit Kundu, Kyle Genova, Xiaoqi Yin, Alireza Fathi, Car- oline Pantofaru, Leonidas J Guibas, Andrea Tagliasacchi, Frank Dellaert, and Thomas Funkhouser. Panoptic Neural Fields: A semantic object-aware neural scene representation. In Proceedings of the IEEE/CVF Conference on...

    2022

  7. [15]

    LiHi-GS: Lidar-supervised Gaussian splat- ting for highway driving scene reconstruction.arXiv preprint arXiv:2412.15447, 2024

    Pou-Chun Kung, Xianling Zhang, Katherine A Skinner, and Nikita Jaipuria. LiHi-GS: Lidar-supervised Gaussian splat- ting for highway driving scene reconstruction.arXiv preprint arXiv:2412.15447, 2024. 1

    2024

  8. [16]

    Urban4d: Semantic-guided 4d gaussian splatting for urban scene reconstruction

    Ziwen Li, Jiaxin Huang, Runnan Chen, Yunlong Che, Yandong Guo, Tongliang Liu, Fakhri Karray, and Ming- ming Gong. Urban4d: Semantic-guided 4d gaussian splatting for urban scene reconstruction. arXiv preprint arXiv:2412.03473, 2024. 2, 3, 4

    2024 arXiv

  9. [17]

    Grounding dino: Marrying dino with grounded pre-training for open-set object detection

    Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Qing Jiang, Chunyuan Li, Jianwei Yang, Hang Su, et al. Grounding dino: Marrying dino with grounded pre-training for open-set object detection. In European Conference on Computer Vision , pages 38–55. Springe...

    2024

  10. [18]

    Smpl: A skinned multi- person linear model

    Matthew Loper, Naureen Mahmood, Javier Romero, Gerard Pons-Moll, and Michael J Black. Smpl: A skinned multi- person linear model. In Seminal Graphics Papers: Pushing the Boundaries, Volume 2, pages 851–866. 2023. 2

    2023

  11. [19]

    Nerf: Representing scenes as neural radiance fields for view syn- thesis

    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. Communications of the ACM , 65(1):99–106, 2021. 2

    2021

  12. [20]

    Instant neural graphics primitives with a mul- tiresolution hash encoding

    Thomas M ¨uller, Alex Evans, Christoph Schied, and Alexan- der Keller. Instant neural graphics primitives with a mul- tiresolution hash encoding. ACM transactions on graphics (TOG), 41(4):1–15, 2022. 2

    2022

  13. [21]

    RoDUS: Ro- bust decomposition of static and dynamic elements in urban scenes

    Thang-Anh-Quan Nguyen, Luis Rold ˜ao, Nathan Piasco, Moussab Bennehar, and Dzmitry Tsishkou. RoDUS: Ro- bust decomposition of static and dynamic elements in urban scenes. In European Conference on Computer Vision, pages 112–130. Springer, 2025. 2, 3, 4

    2025

  14. [22]

    Dinov2: Learning robust visual features without supervision

    Maxime Oquab, Timoth ´ee Darcet, Th ´eo Moutakanni, Huy V o, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, et al. Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193, 2023. 3

    2023 arXiv

  15. [23]

    Neural scene graphs for dynamic scenes

    Julian Ost, Fahim Mannan, Nils Thuerey, Julian Knodt, and Felix Heide. Neural scene graphs for dynamic scenes. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition , pages 2856–2865, 2021. 1, 2

    2021

  16. [24]

    DeSiRe-GS: 4D street Gaussians for static-dynamic decomposition and surface reconstruction for urban driving scenes

    Chensheng Peng, Chengwei Zhang, Yixiao Wang, Chenfeng Xu, Yichen Xie, Wenzhao Zheng, Kurt Keutzer, Masayoshi Tomizuka, and Wei Zhan. DeSiRe-GS: 4D street Gaussians for static-dynamic decomposition and surface reconstruction for urban driving scenes. arXiv preprint arXiv:2411.11921,

    arXiv

  17. [25]

    General matrix representations for B-splines

    Kaihuai Qin. General matrix representations for B-splines. In Proceedings Pacific Graphics’ 98. Sixth Pacific Confer- ence on Computer Graphics and Applications (Cat. No. 98EX208), pages 37–43. IEEE, 1998. 4, 1

    1998

  18. [26]

    Grounded SAM: Assembling open-world models for diverse visual tasks

    Tianhe Ren, Shilong Liu, Ailing Zeng, Jing Lin, Kunchang Li, He Cao, Jiayu Chen, Xinyu Huang, Yukang Chen, Feng Yan, et al. Grounded SAM: Assembling open-world models for diverse visual tasks. arXiv preprint arXiv:2401.14159 ,

    arXiv

  19. [27]

    Structure- from-motion revisited

    Johannes L Schonberger and Jan-Michael Frahm. Structure- from-motion revisited. In Proceedings of the IEEE con- ference on computer vision and pattern recognition , pages 4104–4113, 2016. 1

    2016

  20. [28]

    Very deep convolutional networks for large-scale image recognition

    Karen Simonyan. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556, 2014. 6

    2014 arXiv

  21. [29]

    Scalability in perception for autonomous driving: Waymo open dataset

    Pei Sun, Henrik Kretzschmar, Xerxes Dotiwalla, Aurelien Chouard, Vijaysai Patnaik, Paul Tsui, James Guo, Yin Zhou, Yuning Chai, Benjamin Caine, et al. Scalability in perception for autonomous driving: Waymo open dataset. In Proceed- ings of the IEEE/CVF conference on computer ...

    2020

  22. [30]

    SplatFlow: Self-supervised dynamic Gaus- sian splatting in neural motion flow field for autonomous driving

    Su Sun, Cheng Zhao, Zhuoyang Sun, Yingjie Victor Chen, and Mei Chen. SplatFlow: Self-supervised dynamic Gaus- sian splatting in neural motion flow field for autonomous driving. arXiv preprint arXiv:2411.15482, 2024. 2, 3

    2024 arXiv

  23. [31]

    NeuRAD: Neural rendering for autonomous driving

    Adam Tonderski, Carl Lindstr ¨om, Georg Hess, William Ljungbergh, Lennart Svensson, and Christoffer Petersson. NeuRAD: Neural rendering for autonomous driving. In Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 14895–14904, 2024. 1, 2

    2024

  24. [32]

    SUDS: Scalable urban dynamic scenes

    Haithem Turki, Jason Y Zhang, Francesco Ferroni, and Deva Ramanan. SUDS: Scalable urban dynamic scenes. In Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 12375–12385, 2023. 2, 5, 6, 7

    2023

  25. [33]

    MARS: An instance-aware, modu- lar and realistic simulator for autonomous driving

    Zirui Wu, Tianyu Liu, Liyi Luo, Zhide Zhong, Jianteng Chen, Hongmin Xiao, Chao Hou, Haozhe Lou, Yuantao Chen, Runyi Yang, et al. MARS: An instance-aware, modu- lar and realistic simulator for autonomous driving. In CAAI International Conference on Artificial Intelligence, pages 3–

  26. [34]

    Springer, 2023. 1, 2

    2023

  27. [35]

    Grid4D: 4D decomposed hash encoding for high-fidelity dynamic gaus- sian splatting

    Jiawei Xu, Zexin Fan, Jian Yang, and Jin Xie. Grid4D: 4D decomposed hash encoding for high-fidelity dynamic gaus- sian splatting. In The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024. 6, 7, 8, 1, 2, 3

    2024

  28. [36]

    Street Gaussians: Modeling dynamic urban scenes with Gaussian splatting

    Yunzhi Yan, Haotong Lin, Chenxu Zhou, Weijie Wang, Haiyang Sun, Kun Zhan, Xianpeng Lang, Xiaowei Zhou, and Sida Peng. Street Gaussians: Modeling dynamic urban scenes with Gaussian splatting. In European Conference on Computer Vision, pages 156–173, 2024. 1, 2, 3, 6, 7, 8

    2024

  29. [37]

    Emernerf: Emergent spatial- temporal scene decomposition via self-supervision

    Jiawei Yang, Boris Ivanovic, Or Litany, Xinshuo Weng, Se- ung Wook Kim, Boyi Li, Tong Che, Danfei Xu, Sanja Fidler, Marco Pavone, and Yue Wang. Emernerf: Emergent spatial- temporal scene decomposition via self-supervision. arXiv preprint arXiv:2311.02077, 2023. 2, 6, 8, 1

    2023 arXiv

  30. [38]

    Depth any- thing V2

    Lihe Yang, Bingyi Kang, Zilong Huang, Zhen Zhao, Xiao- gang Xu, Jiashi Feng, and Hengshuang Zhao. Depth any- thing V2. arXiv preprint arXiv:2406.09414, 2024. 5

    2024 arXiv

  31. [39]

    UniSim: A neural closed-loop sensor simulator

    Ze Yang, Yun Chen, Jingkang Wang, Sivabalan Mani- vasagam, Wei-Chiu Ma, Anqi Joyce Yang, and Raquel Ur- tasun. UniSim: A neural closed-loop sensor simulator. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition , pages 1389–1399, 2023. 1, 2

    2023

  32. [40]

    Differentiable surface splatting for point-based geometry processing

    Wang Yifan, Felice Serena, Shihao Wu, Cengiz ¨Oztireli, and Olga Sorkine-Hornung. Differentiable surface splatting for point-based geometry processing. ACM Transactions on Graphics (TOG), 38(6):1–14, 2019. 3

    2019

  33. [41]

    MonoSDF: Exploring monocu- lar geometric cues for neural implicit surface reconstruc- tion

    Zehao Yu, Songyou Peng, Michael Niemeyer, Torsten Sat- tler, and Andreas Geiger. MonoSDF: Exploring monocu- lar geometric cues for neural implicit surface reconstruc- tion. Advances in neural information processing systems , 35:25018–25032, 2022. 5

    2022

  34. [42]

    The unreasonable effectiveness of deep features as a perceptual metric

    Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shecht- man, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. In Proceedings of the IEEE conference on computer vision and pattern recogni- tion, pages 586–595, 2018. 6

    2018

  35. [43]

    HUGS: Holistic urban 3D scene understanding via Gaussian splatting

    Hongyu Zhou, Jiahao Shao, Lu Xu, Dongfeng Bai, Weichao Qiu, Bingbing Liu, Yue Wang, Andreas Geiger, and Yiyi Liao. HUGS: Holistic urban 3D scene understanding via Gaussian splatting. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages ...

    2024

  36. [44]

    DrivingGaussian: Composite Gaussian splatting for surrounding dynamic au- tonomous driving scenes

    Xiaoyu Zhou, Zhiwei Lin, Xiaojun Shan, Yongtao Wang, Deqing Sun, and Ming-Hsuan Yang. DrivingGaussian: Composite Gaussian splatting for surrounding dynamic au- tonomous driving scenes. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages...

    2024

  37. [45]

    car.bus.truck.van.human

    Matthias Zwicker, Hanspeter Pfister, Jeroen Van Baar, and Markus Gross. Surface splatting. In Proceedings of the 28th annual conference on Computer graphics and interac- tive techniques, pages 371–378, 2001. 3 AD-GS: Object-Aware B-Spline Gaussian Splatting for Self-Supervised...

    2001

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.