OccAnyScene: Towards Unified Indoor-Outdoor 3D Occupancy Prediction

Dongmei Jiang; Guiping Cao; Jianguo Zhang; Junjie Liu; Ke Chen; Wanshui Gan; Xiangyuan Lan; Yan Li; Zitong Dai

REVIEW 2 major objections 5 minor 38 references

OccAnyScene: Towards Unified Indoor-Outdoor 3D Occupancy Prediction

T0 review · 2 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read One Gaussian occupancy model can be trained jointly on indoor and outdoor scenes without losing accuracy, because each primitive is sized from the camera-dependent pixel frustum at its predicted depth.

desk verdict A solid, honest occupancy paper with a new task protocol and a good geometric trick, but the headline stability claim needs repeated seeds before it is trusted. read the letter →

arxiv 2608.08696 v2 pith:NHR6IX5N submitted 2026-08-09 cs.CV

Junjie Liu , Wanshui Gan , Zitong Dai , Guiping Cao , Yan Li , Ke Chen , Dongmei Jiang , Xiangyuan Lan

show 1 more author

Jianguo Zhang

This is my paper · ORCID

classification cs.CV

keywords 3Dsemanticoccupancypredictioncross-scenelearningindoorandoutdoorscenesGaussiansplattingpixelfrustumdepthfoundationmodelscene-adaptiveparameterizationvoxel

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

The paper introduces Cross-Scene 3D Semantic Occupancy Prediction, a setting in which a single model must predict occupancy for both indoor and outdoor scenes under their native cameras, spatial ranges, voxel sizes, and semantic label sets. The authors argue that the real obstacle is geometric: Gaussian primitives that represent occupied space must grow with depth and change with focal length, so a shared model cannot rely on absolute metric scales that look nothing alike in a 4.8 m room and a 100 m street. OccAnyScene answers this with a pixel-frustum-centered Gaussian representation whose offsets and scales are expressed relative to the metric cross-section of each pixel's frustum at the predicted depth. The reported results show that one jointly trained model stays within 0.41 mIoU and 0.19 mIoU of separately trained scene-specific models on Occ-ScanNet and SurroundOcc-nuScenes, while a prior Gaussian baseline loses 5.03 and 1.19 points under the same joint training. If correct, this means a general-purpose occupancy predictor can serve environments as different as rooms and roads without per-scene retraining or model switching.

What carries the argument

The central object is the pixel-frustum-centered Gaussian: a 3D Gaussian primitive whose position and scale are constrained by the frustum of the feature pixel that generates it, where a pixel frustum is the pyramid-shaped region of space subtended by that pixel, widening with depth and depending on camera focal length. It is built by two modules: Pixel-Aligned Frustum Feature Aggregation, which combines per-pixel geometry features, the camera ray direction, and surrounding context through deformable cross-attention to produce a frustum query, and Frustum-Parameterized Gaussian Construction, which decodes each query into K Gaussians. The scale of the k-th Gaussian is $s_{p,k} = b_{p,k} \hat{s}_{p,k}$, where $b_{p,k} = \eta \cdot \frac12\,(d_{p,k}/f_x + d_{p,k}/f_y)$ is the metric cross-section width of the pixel frustum at the Gaussian's depth and $\hat{s}_{p,k}$ is a learned dimensionless relative scale. This lets the network predict only depth, bounded offsets, and dimensionless scales, so the metric outcome recomputes itself for any camera intrinsics and scene range.

What would settle it

Retrain the same joint model on a third occupancy dataset with a camera outside the tested range, such as a wide-angle indoor sensor; if the frustum-scale reference drives the stability, the cross-scene mIoU drop should stay near zero, while a miscalibrated reference should reproduce the several-point degradation of the SplatSSC baseline. A cheaper test is to replace $b_{p,k}$ with a single learned constant per dataset and check whether the joint-training gap widens, which would implicate the frustum geometry rather than the depth features.

Watch

Extended reading notes

Core claim

OccAnyScene is a single feed-forward model that predicts 3D semantic occupancy for both indoor Occ-ScanNet and outdoor SurroundOcc-nuScenes protocols, with all weights shared except for dataset-specific taxonomy matrices. It uses a pretrained depth foundation model for visual tokens and dense geometry features, then per feature pixel builds a camera-aware frustum query and decodes it into K Gaussians. The Gaussians are anchored at a predicted metric surface depth, displaced into occluded regions by learned depth increments, shifted laterally by bounded subpixel offsets, and scaled by a dimensionless factor times the frustum's metric width at that depth. The paper reports state-of-the-art scene-specific scores of 59.92% mIoU on Occ-ScanNet and 23.06% mIoU on SurroundOcc-nuScenes, and shows that joint cross-scene training with DepthAnythingV3 costs only 0.41 and 0.19 mIoU respectively, whereas a SplatSSC-based cross-scene baseline drops 5.03 and 1.19 points.

Load-bearing premise

The load-bearing premise is that half the sum of depth over the two focal lengths, the metric width of a pixel's frustum at the Gaussian's depth, is the correct unit of Gaussian size for every camera and scene range, and that its base magnitude, set by an undisclosed constant eta, is calibrated well enough to generalize; if this geometric reference is wrong, the shared parameterization fails exactly in the way the paper claims to prevent.

Editorial extensions

If this is right

A deployed system can serve both road driving and indoor parking with one occupancy model, since only the output taxonomy matrix changes between domains.
Joint training across the two protocols costs only 0.41 and 0.19 mIoU with the DAv3 encoder, indicating that the shared parameterization, not per-scene tuning, is what carries cross-scene stability.
The end-to-end objective optimizes depth and occupancy together in a single loss, removing the two-stage training used by prior Gaussian occupancy methods.
Because the Gaussian scale reference is computed from depth and focal length, the same weights can be evaluated under camera intrinsics different from training without rescaling voxels or re-tuned priors.

Reading between the lines

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

The same frustum-relative argument predicts that the model should transfer to other projection models, such as fisheye or narrow-field cameras, once the frustum cross-section formula is replaced by the appropriate projection geometry; the paper does not test this.
A direct way to separate geometric adaptation from feature transfer is to ablate the frustum reference with a single learned scale per dataset; if the cross-scene gap stays small, the depth foundation model's features, not the frustum geometry, are what make the model stable.
The paper evaluates only two datasets, so the strongest version of the claim, that the parameterization generalizes to arbitrary unseen occupancy protocols, remains open; adding a third indoor or outdoor dataset with a very different camera would be the natural next test.

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.

Desk Editor's Note

read the letter

Two things to know. First, this is not just another occupancy model: it proposes a concrete new task protocol (single model across indoor and outdoor datasets with different camera geometries, voxel specs, and taxonomies) and a clean geometric trick to make it work — expressing Gaussian positions and scales relative to the pixel frustum cross-section at predicted depth. Second, the paper's central quantitative claim, that joint training costs less than half a point of mIoU on each dataset, is plausible but not yet demonstrated because every number in the paper is a single run with no variance.

What is actually new: the pixel-frustum-relative scale formula b = eta*0.5*(d/fx + d/fy) is a sensible, non-obvious way to keep Gaussian sizes comparable across very different cameras and scene ranges. The ablations back it up: removing frustum-relative scale degrades both datasets, and the canonical-camera depth mapping is credited to Metric3Dv2. The paper also does the right thing by wrapping their own SplatSSC baseline with dataset-specific adapters so the cross-scene comparison is apples to apples. The limitations paragraph is unusually candid: it admits the inter-camera gap regions in nuScenes are handled by a separately placed set of spatial queries, which are not described in the method section or Fig. 2 and are never ablated. That is a real exposition gap — if those queries contribute materially to the outdoor numbers, the attribution of stability to the pixel-frustum representation needs to be re-scoped.

The soft spots, in order. First and most important, the two headline drops (0.41 and 0.19 mIoU) sit right in the usual range of seed noise for a model like this. With no error bars, no seed protocol, and no statement about whether the scene-specific and joint runs share initialization, the claim that cross-scene training is 'negligible' is an estimate, not a result. Second, the abstract leads with scene-specific SOTA numbers rather than the cross-scene numbers that are the paper's actual contribution; that's a presentation choice that undercuts the framing but is easy to fix. Third, the inter-camera query module is essentially invisible outside the limitations. Fourth, the hyperparameter eta is introduced but never given a value or sensitivity check; minor.

If I were the editor, I would send this to review. The task setting and parameterization deserve referee time, and the core evidence is consistent even if not yet statistically airtight. I would ask the authors to run multiple seeds (or at least 3) for the scene-specific vs cross-scene comparison, and to fold the spatial-query module into the method and ablations. A reader in 3D occupancy or embodied perception should read this; the task protocol alone is worth knowing about.

Referee Report

2 major / 5 minor

Summary. The paper introduces a new task setting, Cross-Scene 3D Semantic Occupancy Prediction, in which a single model must handle indoor and outdoor scenes with different cameras, spatial ranges, voxel resolutions, and semantic taxonomies. The authors propose OccAnyScene, which builds on a pretrained depth foundation model and consists of two main modules: Pixel-Aligned Frustum Feature Aggregation (PFFA) constructs camera-aware frustum queries, and Frustum-Parameterized Gaussian Construction (FPGC) decodes each query into Gaussians whose positions and scales are constrained by the pixel frustum geometry. The model is evaluated on Occ-ScanNet (indoor) and SurroundOcc-nuScenes (outdoor). The paper reports state-of-the-art scene-specific results and, as its central contribution, shows that a jointly trained cross-scene model nearly matches separately trained scene-specific models, with mIoU drops of 0.41 on Occ-ScanNet and 0.19 on nuScenes using the DAv3 encoder, whereas an adapted SplatSSC baseline drops 5.03 and 1.19 under the same joint training.

Significance. If the central cross-scene stability claim holds, this is a valuable contribution: it defines a meaningful new task and demonstrates a practical way to consolidate indoor and outdoor occupancy prediction into one model without large accuracy loss. The methodological idea of using the pixel frustum as a geometric reference for Gaussian positions and scales is well motivated and is supported by component and fine-grained ablations (Tables 4 and 5), an efficiency analysis, and qualitative visualizations. The paper is also commendably explicit about its limitations, including the dataset scope and the inter-camera coverage mechanism. The main weakness is that the headline quantitative claim rests on single training runs, with no measure of run-to-run variability; for a claim of 'negligible degradation', that evidence is not yet conclusive.

major comments (2)

[Section 4.3, Tables 2 and 3] The central claim that joint cross-scene training causes negligible degradation is based on mIoU differences of 0.41 on Occ-ScanNet and 0.19 on SurroundOcc-nuScenes between the scene-specific and cross-scene models. The paper reports only one training run per configuration, with no standard deviation, no number of seeds, and no statement about whether the scene-specific and cross-scene runs shared initialization or hyperparameters. On a metric like mIoU, run-to-run variation is commonly of this magnitude, so the reported stability is not yet distinguished from noise. Please provide repeated-seed results (at least three runs, with means and standard deviations) for the scene-specific versus cross-scene comparison, or otherwise quantify the expected variability of the differences.
[Section 5 (Limitations) vs. Section 3 and Fig. 2] The model includes a supplementary spatial-query module that covers inter-camera gaps in SurroundOcc-nuScenes, but this module is described only in the Limitations section and is absent from the method description and Fig. 2. No ablation quantifies its effect on nuScenes metrics. Since the paper attributes the stability of the cross-scene model to the pixel-frustum representation, and the queries produce the occupancy predictions in camera-unobserved regions, their contribution to the reported nuScenes numbers should be measured. The statement that these queries have little effect on aggregate metrics is an assertion, not evidence; please provide an ablation with and without the module, and if the module is retained, document it in the method section.

minor comments (5)

[Abstract and Section 4.3] The abstract highlights the scene-specific SOTA numbers (59.92% and 23.06% mIoU) rather than the cross-scene stability numbers, which are the paper's novel contribution. It would be clearer to state the cross-scene results and their near-equivalence to scene-specific models in the abstract.
[Table 2, row OccAnyScene-DAv3 Cross-scene] The entry '59.1064.40' appears to be a typographical error and likely should be '59.10 64.40' with a separator; the current formatting makes the table difficult to read.
[Section 4.4, Table 5] The text refers to 'depth increments' while the table row is labeled 'w/o depth residual'; please use consistent terminology for this component.
[Section 3.3, pixel-frustum-relative scale parameterization] The scale reference b_{p,k} = eta * 0.5 * (d/fx + d/fy) is the key mechanism for cross-scene scale adaptation, but the value of the hyperparameter eta is not reported and no sensitivity analysis is provided. Please state the chosen value and, ideally, show how results vary with eta.
[Section 4.2] The implementation details state that the effective number of iterations per dataset is kept identical to the single-dataset setting, but no details are given on learning rate schedules, batch sizes, or how the alternating schedule is implemented; please clarify these for reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the cross-scene stability claim is an empirical benchmark comparison, with no fitted parameter renamed as prediction and no load-bearing self-citation chain.

full rationale

The paper's central claim is empirical: a single model jointly trained on Occ-ScanNet and SurroundOcc-nuScenes achieves mIoU comparable to separately trained scene-specific models (drops of 0.41 and 0.19 with DAv3). The method chain consists of a pretrained depth foundation model, Pixel-Aligned Frustum Feature Aggregation, Frustum-Parameterized Gaussian Construction, and Gaussian-to-voxel splatting using SplatSSC's Decoupled Gaussian Aggregator. No parameter is fitted to the target metric and then reported as a prediction; the frustum-relative scale reference b_{p,k} = eta * 0.5 * (d/fx + d/fy) is a geometric formula with a fixed hyperparameter eta, and Table 5 shows that removing it degrades both datasets, so it is empirically motivated rather than co-constructed with the result. Canonical-camera depth follows Metric3Dv2 (external) and DepthAnythingV2/V3 (external pretrained encoders). The self-citations [5] and [6] appear only in the related-work survey of rendering-based supervision and are not load-bearing for any derivation. The paper's own Limitations section concedes that evaluation covers only two datasets rather than arbitrary unseen scenes and that inter-camera gaps are filled by an unablated supplementary spatial-query module absent from Section 3; this is a completeness and reproducibility concern, not a circular reduction. The single-run, no-seed comparison underlying the small mIoU drops is a statistical robustness risk, but that too is distinct from circularity. No derivation in the paper reduces by construction to its own inputs.

Assumptions & free parameters 4 free parameters · 4 assumptions · 1 invented entities

The central contribution is empirical and builds on several external components. The paper contributes the pixel-frustum parameterization but relies on pretrained depth features, canonical depth scaling, learned taxonomy matrices, and an external Gaussian-to-voxel aggregator. The free parameters are mostly hyperparameters left undisclosed; the invented entity is the supplementary Gaussian queries for inter-camera gaps.

free parameters (4)

eta (frustum scale base magnitude) = not disclosed
Fixed hyperparameter controlling the base magnitude of the pixel-frustum scale reference in Section 3.3; chosen by hand and not reported.
K (Gaussians per pixel frustum) = 3 (default)
Number of Gaussians decoded per pixel frustum; ablated in Table 6, with K=1 already competitive.
Supplementary spatial query count and placement = not specified
Learnable query embeddings placed in camera-unobserved areas for nuScenes; configuration not explained in the paper.
Loss weights (lambda_focal, lambda_lov, lambda_scal, lambda_depth) = not reported
Weights for the occupancy and depth losses in Section 3.4 are not given numerically.

assumptions (4)

domain assumption Pretrained depth foundation model (DepthAnythingV2/V3) provides dense geometry features and visual tokens that transfer across indoor and outdoor domains.
The method is built on these features without fine-tuning the foundation model, assuming their cross-domain quality (Section 3.2).
domain assumption Canonical camera depth scaling from Metric3Dv2, D_surf = D_can * f_real / f_can, yields metric surface depth across different camera intrinsics.
Invoked in Section 3.3 to make depth anchors metric-consistent across datasets.
ad hoc to paper The pixel-frustum cross-section formula b_{p,k} = eta * 0.5 * (d/fx + d/fy) is a valid geometric scale reference for Gaussian size.
This is the paper's key design choice for scene-adaptive scales; it is plausible but not derived from a theory, and eta is undisclosed (Section 3.3).
domain assumption The Decoupled Gaussian Aggregator from SplatSSC effectively maps Gaussian primitives to arbitrary voxel grids.
Reused as the splatting module without modification (Section 3.4), assuming its correctness and generality.

invented entities (1)

Supplementary spatial queries and their decoded Gaussians
purpose: To cover inter-camera gaps in SurroundOcc-nuScenes that pixel frusta cannot reach, by cross-attending to pooled image features and decoding into Gaussian primitives.
Described only in the Limitations section, their placement and impact on the nuScenes metrics are not ablated or externally validated.

how reviews work

0 comments

Cite this review

Pith. "Pith review of OccAnyScene: Towards Unified Indoor-Outdoor 3D Occupancy Prediction." pith.science (2026). https://pith.science/paper/NHR6IX5N

@misc{pith2026260808696,
  author       = {Pith},
  title        = {Pith review of: OccAnyScene: Towards Unified Indoor-Outdoor 3D Occupancy Prediction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NHR6IX5N}},
  note         = {Machine review of arXiv:2608.08696}
}

read the original abstract

3D occupancy prediction is fundamental to scene understanding, yet existing 3D semantic occupancy methods are typically specialized to fixed scene types and occupancy protocols. We introduce Cross-Scene 3D Semantic Occupancy Prediction, a new task setting which requires a single model to handle heterogeneous indoor and outdoor scenes with varying cameras, spatial ranges, voxel specifications, and semantic taxonomies. This setting poses a fundamental challenge: achieving metric-consistent yet scene-adaptive image-to-3D lifting across varying camera configurations and scene scales. To address this challenge, we propose OccAnyScene, a pixel-frustum-centered Gaussian framework built upon a pretrained depth foundation model. Specifically, the framework employs Pixel-Aligned Frustum Feature Aggregation to construct a camera-aware frustum query for each feature pixel, and Frustum-Parameterized Gaussian Construction to decode each query into multiple Gaussians whose positions and sizes are constrained by the predicted pixel depth and corresponding frustum geometry. OccAnyScene sets new state-of-the-art results, achieving 59.92% mIoU on the indoor Occ-ScanNet and 23.06% mIoU on the outdoor SurroundOcc-nuScenes.

Figures

Figures reproduced from arXiv: 2608.08696 by the authors.

**Figure 2.** Overview of OccAnyScene, a pixel-frustum-centered Gaussian framework. OccAnyScene treats each pixel frustum as the [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** Qualitative results on Occ-ScanNet and SurroundOcc-nuScenes. We compare scene-specific and cross-scene OccAnyScene [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Figures from the paper (4 more)

**Figure 4.** Figure 4: Per-Gaussian depth increments on Occ-ScanNet. With [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

**Figure 5.** Figure 5: Efficiency comparison on Occ-ScanNet in terms of pa [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

**Figure 6.** Figure 6: Occlusion-aware Gaussian positioning on SurroundOcc-nuScenes. Results are produced by the jointly trained cross-scene OccAnyScene model. In the left panel, the six columns correspond to the six camera views, while the rows show (a) input images, (b) predicted surface d…

**Figure 7.** Figure 7: Occlusion-aware Gaussian positioning on OccScanNet. The rows show (a) the input image, (b) predicted surface depth, (c) the depth of the Gaussians decoded from each pixel frustum, (d) the surface-relative depth increment, and (e) the final semantic occupancy predicti…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 21 canonical work pages

[1]

Omni3d: A large benchmark and model for 3d object detection in the wild

Garrick Brazil, Abhinav Kumar, Julian Straub, Nikhila Ravi, Justin Johnson, and Georgia Gkioxari. Omni3d: A large benchmark and model for 3d object detection in the wild. InProceedings of the IEEE/CVF conference on computer vi- sion and pattern recognition, pages 13154–13164, 2023

work page 2023
[2]

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. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11621–11631, 2020

work page 2020
[3]

Monoscene: Monoc- ular 3d semantic scene completion

Anh-Quan Cao and Raoul De Charette. Monoscene: Monoc- ular 3d semantic scene completion. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 3991–4001, 2022

2022
[4]

Occany: Generalized unconstrained urban 3d occupancy

Anh-Quan Cao and Tuan-Hung Vu. Occany: Generalized unconstrained urban 3d occupancy. InConference on Com- puter Vision and Pattern Recognition (CVPR), 2026

work page 2026
[5]

A comprehensive framework for 3d occupancy estimation in autonomous driving.IEEE Transactions on Intelligent Vehi- cles, 2024

Wanshui Gan, Ningkai Mo, Hongbin Xu, and Naoto Yokoya. A comprehensive framework for 3d occupancy estimation in autonomous driving.IEEE Transactions on Intelligent Vehi- cles, 2024

work page 2024
[6]

Gaussianocc: Fully self-supervised and ef- ficient 3d occupancy estimation with gaussian splatting

Wanshui Gan, Fang Liu, Hongbin Xu, Ningkai Mo, and Naoto Yokoya. Gaussianocc: Fully self-supervised and ef- ficient 3d occupancy estimation with gaussian splatting. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 28980–28990, 2025

work page 2025
[7]

3d gaussian representations in semantic occupancy prediction: A comprehensive survey and analysis

Seamie Hayes, Reenu Mohandas, Tim Brophy, Ganesh Sistu, and Ciaran Eising. 3d gaussian representations in semantic occupancy prediction: A comprehensive survey and analysis. Authorea Preprints, 2025

work page 2025
[8]

Mu Hu, Wei Yin, Chi Zhang, Zhipeng Cai, Xiaoxiao Long, Hao Chen, Kaixuan Wang, Gang Yu, Chunhua Shen, and Shaojie Shen. Metric3d v2: A versatile monocular geomet- ric foundation model for zero-shot metric depth and surface normal estimation.IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(12):10579–10596, 2024

2024

Show all 38 references

[9]

Tri-perspective view for vision- based 3d semantic occupancy prediction

Yuanhui Huang, Wenzhao Zheng, Yunpeng Zhang, Jie Zhou, and Jiwen Lu. Tri-perspective view for vision- based 3d semantic occupancy prediction. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 9223–9232, 2023

2023
[10]

Gaussianformer: Scene as gaussians for vision-based 3d semantic occupancy prediction

Yuanhui Huang, Wenzhao Zheng, Yunpeng Zhang, Jie Zhou, and Jiwen Lu. Gaussianformer: Scene as gaussians for vision-based 3d semantic occupancy prediction. InEuropean Conference on Computer Vision, pages 376–393. Springer, 2024

2024
[11]

Gaussianformer-2: Probabilistic gaussian superposition for efficient 3d occupancy prediction

Yuanhui Huang, Amonnut Thammatadatrakoon, Wenzhao Zheng, Yunpeng Zhang, Dalong Du, and Jiwen Lu. Gaussianformer-2: Probabilistic gaussian superposition for efficient 3d occupancy prediction. InProceedings of the computer vision and pattern recognition conference, pages 27477–2...

2025
[12]

3d gaussian splatting for real-time radiance field rendering.ACM Transactions on Graphics, 42 (4), 2023

Bernhard Kerbl, Georgios Kopanas, Thomas Leimk ¨uhler, and George Drettakis. 3d gaussian splatting for real-time radiance field rendering.ACM Transactions on Graphics, 42 (4), 2023

2023
[13]

V oxformer: Sparse voxel transformer for camera- based 3d semantic scene completion

Yiming Li, Zhiding Yu, Christopher Choy, Chaowei Xiao, Jose M Alvarez, Sanja Fidler, Chen Feng, and Anima Anand- kumar. V oxformer: Sparse voxel transformer for camera- based 3d semantic scene completion. InProceedings of the IEEE/CVF conference on computer vision and pattern ...

2023
[14]

Depth anything 3: Recovering the visual space from any views

Haotong Lin, Sili Chen, Junhao Liew, Donny Y Chen, Zhenyu Li, Guang Shi, Jiashi Feng, and Bingyi Kang. Depth anything 3: Recovering the visual space from any views. arXiv preprint arXiv:2511.10647, 2025

2025 arXiv
[15]

Fully sparse 3d occupancy prediction

Haisong Liu, Yang Chen, Haiguang Wang, Zetong Yang, Tianyu Li, Jia Zeng, Li Chen, Hongyang Li, and Limin Wang. Fully sparse 3d occupancy prediction. InEuropean Conference on Computer Vision, pages 54–71. Springer, 2024

2024
[16]

Vision-centric bev perception: A survey.IEEE Trans- actions on Pattern Analysis and Machine Intelligence, 46 (12):10978–10997, 2024

Yuexin Ma, Tai Wang, Xuyang Bai, Huitong Yang, Yuenan Hou, Yaming Wang, Yu Qiao, Ruigang Yang, and Xinge Zhu. Vision-centric bev perception: A survey.IEEE Trans- actions on Pattern Analysis and Machine Intelligence, 46 (12):10978–10997, 2024

2024
[17]

Splatssc: Decoupled depth-guided gaussian splatting for semantic scene completion

Rui Qian, Haozhi Cao, Tianchen Deng, Shenghai Yuan, and Lihua Xie. Splatssc: Decoupled depth-guided gaussian splatting for semantic scene completion. InProceedings of the AAAI Conference on Artificial Intelligence, pages 8520– 8528, 2026

2026
[18]

Robofactory: Exploring embodied agent collaboration with compositional constraints

Yiran Qin, Li Kang, Xiufeng Song, Zhenfei Yin, Xiaohong Liu, Xihui Liu, Ruimao Zhang, and Lei Bai. Robofactory: Exploring embodied agent collaboration with compositional constraints. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 10075–10085, 2025

2025
[19]

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. InProceed- ings of the IEEE/CVF conference on computer v...

2020
[20]

Splatter image: Ultra-fast single-view 3d recon- struction

Stanislaw Szymanowicz, Chrisitian Rupprecht, and Andrea Vedaldi. Splatter image: Ultra-fast single-view 3d recon- struction. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10208– 10217, 2024

2024
[21]

Flash3d: Feed-forward general- isable 3d scene reconstruction from a single image

Stanislaw Szymanowicz, Eldar Insafutdinov, Chuanxia Zheng, Dylan Campbell, Joao F Henriques, Christian Rup- precht, and Andrea Vedaldi. Flash3d: Feed-forward general- isable 3d scene reconstruction from a single image. In2025 International Conference on 3D Vision (3DV), pages 670–
[22]

Occ3d: A large-scale 3d occupancy prediction benchmark for autonomous driving.Advances in Neural Information Processing Systems, 36:64318–64330, 2023

Xiaoyu Tian, Tao Jiang, Longfei Yun, Yucheng Mao, Huitong Yang, Yue Wang, Yilun Wang, and Hang Zhao. Occ3d: A large-scale 3d occupancy prediction benchmark for autonomous driving.Advances in Neural Information Processing Systems, 36:64318–64330, 2023

2023
[23]

Embodiedocc++: Boosting embodied 3d 10 occupancy prediction with plane regularization and uncer- tainty sampler

Hao Wang, Xiaobao Wei, Xiaoan Zhang, Jianing Li, Chengyu Bai, Ying Li, Ming Lu, Wenzhao Zheng, and Shanghang Zhang. Embodiedocc++: Boosting embodied 3d 10 occupancy prediction with plane regularization and uncer- tainty sampler. InProceedings of the 33rd ACM Interna- tional Co...

2025
[24]

Embodiedscan: A holistic multi- modal 3d perception suite towards embodied ai

Tai Wang, Xiaohan Mao, Chenming Zhu, Runsen Xu, Ruiyuan Lyu, Peisen Li, Xiao Chen, Wenwei Zhang, Kai Chen, Tianfan Xue, et al. Embodiedscan: A holistic multi- modal 3d perception suite towards embodied ai. InProceed- ings of the IEEE/CVF Conference on Computer Vision and Patte...

2024
[25]

Openoccupancy: A large scale benchmark for surrounding semantic occupancy perception

Xiaofeng Wang, Zheng Zhu, Wenbo Xu, Yunpeng Zhang, Yi Wei, Xu Chi, Yun Ye, Dalong Du, Jiwen Lu, and Xin- gang Wang. Openoccupancy: A large scale benchmark for surrounding semantic occupancy perception. InProceedings of the IEEE/CVF International Conference on Computer Vi- sion...

2023
[26]

Surroundocc: Multi-camera 3d occu- pancy prediction for autonomous driving

Yi Wei, Linqing Zhao, Wenzhao Zheng, Zheng Zhu, Jie Zhou, and Jiwen Lu. Surroundocc: Multi-camera 3d occu- pancy prediction for autonomous driving. InProceedings of the IEEE/CVF International Conference on Computer Vi- sion, pages 21729–21740, 2023

2023
[27]

Embodiedocc: Embodied 3d occupancy prediction for vision-based online scene understanding

Yuqi Wu, Wenzhao Zheng, Sicheng Zuo, Yuanhui Huang, Jie Zhou, and Jiwen Lu. Embodiedocc: Embodied 3d occupancy prediction for vision-based online scene understanding. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 26360–26370, 2025

2025
[28]

A survey on occupancy perception for au- tonomous driving: The information fusion perspective.In- formation Fusion, 114:102671, 2025

Huaiyuan Xu, Junliang Chen, Shiyu Meng, Yi Wang, and Lap-Pui Chau. A survey on occupancy perception for au- tonomous driving: The information fusion perspective.In- formation Fusion, 114:102671, 2025

2025
[29]

Vg3s: Vi- sual geometry grounded gaussian splatting for semantic oc- cupancy prediction.arXiv preprint arXiv:2603.06210, 2026

Xiaoyang Yan, Muleilan Pei, and Shaojie Shen. Vg3s: Vi- sual geometry grounded gaussian splatting for semantic oc- cupancy prediction.arXiv preprint arXiv:2603.06210, 2026

2026 arXiv
[30]

Depth any- thing v2.Advances in Neural Information Processing Sys- tems, 37:21875–21911, 2024

Lihe Yang, Bingyi Kang, Zilong Huang, Zhen Zhao, Xiao- gang Xu, Jiashi Feng, and Hengshuang Zhao. Depth any- thing v2.Advances in Neural Information Processing Sys- tems, 37:21875–21911, 2024

2024
[31]

Monocular occupancy prediction for scalable indoor scenes

Hongxiao Yu, Yuqi Wang, Yuntao Chen, and Zhaoxiang Zhang. Monocular occupancy prediction for scalable indoor scenes. InEuropean Conference on Computer Vision, pages 38–54. Springer, 2024

2024
[32]

Flashocc: Fast and memory-efficient occupancy prediction via channel-to-height plugin.arXiv preprint arXiv:2311.12058, 2023

Zichen Yu, Changyong Shu, Jiajun Deng, Kangjie Lu, Zong- dai Liu, Jiangyong Yu, Dawei Yang, Hui Li, and Yan Chen. Flashocc: Fast and memory-efficient occupancy prediction via channel-to-height plugin.arXiv preprint arXiv:2311.12058, 2023

2023 arXiv
[33]

Context and geometry aware voxel transformer for semantic scene completion.Advances in Neural Information Processing Systems, 37:1531–1555, 2024

Zhu Yu, Runmin Zhang, Jiacheng Ying, Junchen Yu, Xiaohai Hu, Lun Luo, Si-Yuan Cao, and Hui-Liang Shen. Context and geometry aware voxel transformer for semantic scene completion.Advances in Neural Information Processing Systems, 37:1531–1555, 2024

2024
[34]

Occformer: Dual-path transformer for vision-based 3d semantic occu- pancy prediction

Yunpeng Zhang, Zheng Zhu, and Dalong Du. Occformer: Dual-path transformer for vision-based 3d semantic occu- pancy prediction. InProceedings of the IEEE/CVF Inter- national Conference on Computer Vision, pages 9433–9443, 2023

2023
[35]

Roboocc: Enhancing the geometric and semantic scene understanding for robots.arXiv preprint arXiv:2504.14604, 2025

Zhang Zhang, Qiang Zhang, Wei Cui, Shuai Shi, Yijie Guo, Gang Han, Wen Zhao, Hengle Ren, Renjing Xu, and Jian Tang. Roboocc: Enhancing the geometric and semantic scene understanding for robots.arXiv preprint arXiv:2504.14604, 2025

2025 arXiv
[36]

Generalizing visual geometry pri- ors to sparse gaussian occupancy prediction

Changqing Zhou, Yueru Luo, Yulan Guo, Bing Wang, Jie Qin, and Changhao Chen. Generalizing visual geometry pri- ors to sparse gaussian occupancy prediction. InProceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), 2026

2026
[37]

Dlwm: Dual latent world models enable holistic gaussian-centric pre-training in autonomous driving

Yiyao Zhu, Ying Xue, Haiming Zhang, Guangfeng Jiang, Wending Zhou, Xu Yan, Jiantao Gao, Yingjie Cai, Bingbing Liu, Zhen Li, et al. Dlwm: Dual latent world models enable holistic gaussian-centric pre-training in autonomous driving. InProceedings of the IEEE/CVF Conference on Co...

2026
[38]

Gaussianworld: Gaussian world model for streaming 3d occupancy prediction

Sicheng Zuo, Wenzhao Zheng, Yuanhui Huang, Jie Zhou, and Jiwen Lu. Gaussianworld: Gaussian world model for streaming 3d occupancy prediction. InProceedings of the Computer Vision and Pattern Recognition Conference, pages 6772–6781, 2025. 11

2025

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Omni3d: A large benchmark and model for 3d object detection in the wild

Garrick Brazil, Abhinav Kumar, Julian Straub, Nikhila Ravi, Justin Johnson, and Georgia Gkioxari. Omni3d: A large benchmark and model for 3d object detection in the wild. InProceedings of the IEEE/CVF conference on computer vi- sion and pattern recognition, pages 13154–13164, 2023

work page 2023

[2] [2]

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. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11621–11631, 2020

work page 2020

[3] [3]

Monoscene: Monoc- ular 3d semantic scene completion

Anh-Quan Cao and Raoul De Charette. Monoscene: Monoc- ular 3d semantic scene completion. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 3991–4001, 2022

2022

[4] [4]

Occany: Generalized unconstrained urban 3d occupancy

Anh-Quan Cao and Tuan-Hung Vu. Occany: Generalized unconstrained urban 3d occupancy. InConference on Com- puter Vision and Pattern Recognition (CVPR), 2026

work page 2026

[5] [5]

A comprehensive framework for 3d occupancy estimation in autonomous driving.IEEE Transactions on Intelligent Vehi- cles, 2024

Wanshui Gan, Ningkai Mo, Hongbin Xu, and Naoto Yokoya. A comprehensive framework for 3d occupancy estimation in autonomous driving.IEEE Transactions on Intelligent Vehi- cles, 2024

work page 2024

[6] [6]

Gaussianocc: Fully self-supervised and ef- ficient 3d occupancy estimation with gaussian splatting

Wanshui Gan, Fang Liu, Hongbin Xu, Ningkai Mo, and Naoto Yokoya. Gaussianocc: Fully self-supervised and ef- ficient 3d occupancy estimation with gaussian splatting. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 28980–28990, 2025

work page 2025

[7] [7]

3d gaussian representations in semantic occupancy prediction: A comprehensive survey and analysis

Seamie Hayes, Reenu Mohandas, Tim Brophy, Ganesh Sistu, and Ciaran Eising. 3d gaussian representations in semantic occupancy prediction: A comprehensive survey and analysis. Authorea Preprints, 2025

work page 2025

[8] [8]

Mu Hu, Wei Yin, Chi Zhang, Zhipeng Cai, Xiaoxiao Long, Hao Chen, Kaixuan Wang, Gang Yu, Chunhua Shen, and Shaojie Shen. Metric3d v2: A versatile monocular geomet- ric foundation model for zero-shot metric depth and surface normal estimation.IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(12):10579–10596, 2024

2024