Pith. sign in

REVIEW 4 major objections 6 minor 43 references

PVP: Polar Representation Boost for 3D Semantic Occupancy Prediction

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read Polar voxel grids, with distortion-correcting modules, outdo Cartesian methods for 3D occupancy prediction.

desk verdict Solid incremental architecture with strong claimed gains on OpenOccupancy, but the SOTA margins rest on unverified baseline numbers and no code; worth refereeing but require reproducibility materials. read the letter →

arxiv 2412.07616 v2 pith:FZCEL3U2 submitted 2024-12-10 cs.CV

Yujing Xue , Jiaxiang Liu , Jiawei Du , Joey Tianyi Zhou This is my paper · ORCID
classification cs.CV
keywords 3DsemanticoccupancypredictionpolarrepresentationfeaturedistortionattentionmechanismplanedecomposedconvolutionLiDAR-camerafusionautonomousdriving
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

This paper argues that representing the surrounding 3D space in polar coordinates, with finer voxels near the sensor and coarser voxels far away, can beat the standard Cartesian voxel grid for semantic occupancy prediction provided the distortions polar grids introduce are explicitly handled. To that end it proposes PVP, built from two modules: GRP, an attention mechanism that propagates features along the radial, azimuth, and height axes so large structures like roads stay coherent, and PD-Conv, which replaces full 3D convolutions with three plane-wise 2D convolutions that match how distortion actually varies across views. On the OpenOccupancy benchmark, PVP reports a 1.9% mIoU and 2.9% IoU gain over the best LiDAR-only method and 7.9% mIoU and 6.8% IoU over the best multimodal baseline. A sympathetic reader would take the paper's central claim to be that polar representation, once its distortions are corrected, is not merely viable but superior for dense 3D scene understanding.

What carries the argument

The load-bearing machinery is the pair of distortion-countering modules. GRP first applies local condense attention, which selects a representative feature per non-overlapping window and refines it with cross-attention against the window's own features, producing a condensed volume; then global decomposed attention runs self-attention in strips along the radial, azimuth, and height axes in turn, so distant parts of a road or building can interact at low cost, and finally a reverse cross-attention propagates the calibrated features back to the full resolution. PD-Conv replaces each 3x3x3 convolution with three stacked decomposed convolutions (1x3x3, 3x1x3, 3x3x1), each acting on one plane of the polar volume, which separates the scale, projection, and identity distortions onto different planes. The final output is resampled to Cartesian voxels via trilinear sampling and passed through the same occupancy head, losses, and coarse-to-fine refinement as the OpenOccupancy baseline.

What would settle it

Rerun PointOcc and M-CONet with the exact training schedule (25 epochs, batch size 8, V100 GPUs), voxel range [0.3, 73]m on the rho axis, and the same evaluation code; if the mIoU gaps over these baselines shrink to well below the reported 1.9% and 7.9%, the central claim of state-of-the-art performance is refuted.

Watch

Extended reading notes

Core claim

PVP is a 3D semantic occupancy predictor that operates on a polar voxel grid instead of a Cartesian one. The paper's central claim is that the two design elements it introduces resolve the two failure modes that have held polar representations back in dense prediction: GRP (Global Representation Propagation) fixes the global misalignment of large 'stuff' classes by condensing local windows into representative features, running decomposed attention along each axis, and propagating the corrected features back to every voxel; PD-Conv (Plane Decomposed Convolution) fixes local distortion by replacing each 3D convolution with three 2D convolutions, a scale transformation on the range plane, a projection transformation on the BEV plane, and an identity transformation on the slicing plane. With these changes, PVP reports the strongest results on the OpenOccupancy validation set, improving IoU to 37.0% (LiDAR-only) and 36.3% (camera+LiDAR) and mIoU to 25.8% and 28.0%, respectively, surpassing the previous state of the art in each input modality.

Load-bearing premise

The reported improvements assume the published baseline results are directly comparable; the paper aligns pipelines with the benchmark but does not rerun the baselines or verify identical training schedules, voxelization ranges, and evaluation code.

Editorial extensions

If this is right

  • If the reported numbers hold, polar-coordinate voxelization becomes the recommended representation for 3D semantic occupancy prediction in autonomous driving, because it spends resolution where sensors actually provide information.
  • PVP's gains are largest on stuff classes such as roads, sidewalks, and terrain, suggesting the approach specifically repairs the failure mode that Cartesian methods share for large, elongated structures.
  • The method keeps memory use near or below the Cartesian baseline while improving accuracy, per the paper's efficiency analysis, making it viable for onboard deployment.
  • The polar-to-Cartesian trilinear sampling at the output head means downstream consumers of the predictions still receive a standard Cartesian voxel grid.

Reading between the lines

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

  • Because the baselines are quoted from published results rather than re-run, the exact margins of 1.9/2.9 and 7.9/6.8 percentage points should be treated as conditional on protocol alignment until an independent re-implementation confirms them.
  • The observation that reordering the three decomposed convolutions barely changes performance hints that the benefit comes from separating the axes, not from the specific receptive field; a testable extension is to use randomly permuted kernel orders and measure the variance.
  • GRP's design could transfer to other dense prediction tasks dominated by large background classes (e.g., BEV map segmentation), where global feature propagation along structured axes may repair similar distortion failures.
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

4 major / 6 minor

Summary. The paper proposes PVP, a 3D semantic occupancy predictor that operates in polar coordinates rather than the usual Cartesian volume. It introduces two components: a Global Represent Propagation (GRP) module that condenses local windows and then performs axis-wise global attention to propagate features, and a Plane Decomposed Convolution (PD-Conv) that replaces 3D convolutions with three 2D convolutions on different planes. The model is built on the CONet/OpenOccupancy pipeline, loss, and prediction head. On the OpenOccupancy (nuScenes) validation set, the authors report that LiDAR-only PVP exceeds PointOcc by 1.9 mIoU and 2.9 IoU, and that multimodal PVP exceeds M-CONet by 7.9 mIoU and 6.8 IoU. Component ablations at 5 epochs show consistent gains for GRP and PD-Conv over a polar-only baseline. The paper does not release code, logs, or rerun baseline methods; all comparisons rely on published numbers.

Significance. If the reported empirical gains are reproducible under a fair protocol, the polar-coordinate design is a meaningful step for 3D semantic occupancy prediction, and the two proposed modules address a real distortion problem. The paper's internal ablations are systematic and show that the full model is much stronger than a direct polar conversion, which is a useful finding. However, the central SOTA claim rests entirely on baseline numbers taken from prior papers without any evidence that training schedules, voxelization ranges, loss weights, or evaluation code match. The absence of error bars, multi-seed runs, and code further weakens the quantitative claims. The contribution is therefore defensible in design but not yet supported in its headline comparison.

major comments (4)
  1. [4.1.2, Table 1] The paper's central claim—surpassing PointOcc by +1.9 mIoU/+2.9 IoU and M-CONet by +7.9 mIoU/+6.8 IoU—is supported only by published baseline numbers. The sentence in Section 4.1.2, 'All baseline implementations were aligned with the benchmark [16] to ensure fairness in comparison,' is not backed by any rerun, training log, configuration file, or evaluation script. Small protocol differences (number of epochs, learning-rate schedule, voxelization range, polar-to-Cartesian sampling, loss weighting) can plausibly shift mIoU by more than a point. Since Table 1 is the only evidence for the SOTA claim, the authors need to provide either rerun baselines under the exact PVP protocol or explicitly state that the compared numbers are quoted from the original papers and soften the 'surpasses' wording accordingly.
  2. [4.4, Tables 2, 3, 5] All ablations are single-run results at 5 epochs, while the headline results in Table 1 are at 25 epochs. There is no evidence that the 5-epoch ranking is representative of the 25-epoch setting. In Table 3, four variants (Channel, Swin, GRR, and Polar-no-attention) are separated by only 0.1 mIoU, which is almost certainly within run-to-run noise. The claims that GRP is better than these alternatives, and that the PD-Conv structure variations are negligible, are therefore underdetermined. The authors should provide error bars or multiple seeds for the key rows, or run the full 25-epoch protocol at least for the main ablation path.
  3. [3.2.1, Eq. (2); 3.2.2] The GRP module is a core contribution, but Eq. (2) is not a well-formed attention operation. It is written as `softmax( Q_i K_i / sqrt(d) · V_i + E(p) )`, which appears to multiply the attention weights by V without summing over key positions, and it adds positional encoding after the dot product rather than inside the softmax. The `maxsel` operation is not formally defined, the relation between the window size `Wg` and the downsampling factor `S` in the condensed feature map is never stated, and the reverse cross attention in Section 3.2.2 uses an undefined dimension `N`. As written, this part of the method cannot be reproduced or checked.
  4. [3.3, 4.2] PD-Conv, the second core contribution, is under-specified. The text says the three decomposed blocks are stacked in series and then combined in parallel, but it does not give the exact block order, the number of blocks, kernel sizes, strides, padding, how the parallel branches are aggregated, or which stages of the 3D backbone are replaced. The statement that 'the stacking order does not significantly affect performance' is qualitative and is only validated at 5 epochs. Without a precise specification, the method is not reproducible from the paper, and the claimed source of the gain (plane decomposition rather than asymmetric kernel layout) cannot be independently verified.
minor comments (6)
  1. [Table 4] The rows labeled 'OpenOccupancy' report IoU/mIoU of 30.7/15.0 (L) and 29.3/19.8 (C&L), while Table 1 lists L-CONet as 30.9/15.8 and M-CONet as 29.5/20.1. Please clarify whether 'OpenOccupancy' and 'CONet' refer to the same implementation and why the numbers differ; otherwise the efficiency comparison is ambiguous.
  2. [Abstract, Section 4.3] The abstract says 'LiDAR-only methods' and 'multimodal methods' without naming the specific baselines; the actual comparison is against PointOcc and M-CONet. Please name the strongest prior method in each category to make the claim precise.
  3. [4.1.1] The dataset description says '32-lane LiDAR'; this should be '32-beam LiDAR'.
  4. [3.2.1] There is a likely typo in Eq. (2): 'sof tmax' should be 'softmax', and the inner expression lacks the conventional scaling/summation notation. Please rewrite the equation in standard form.
  5. [Figure 3] The Figure 3 caption is incomplete: it ends without a period and contains 'The pipeline ofPVP'. Please proofread the captions and the main text for similar typographical issues (e.g., 'tenser', '3D V oxel-based').
  6. [Section 4.1.2] The list of compared baselines does not mention the training schedule or evaluation protocol used for each published baseline. Even if the numbers are quoted, a sentence stating that no baseline was rerun would help readers assess the fairness claim.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: PVP's claims are empirical gains on an external benchmark; self-citations to PARTNER are design antecedents, not load-bearing derivations.

full rationale

The derivation chain in PVP is architectural, not analytical: the paper proposes GRP and PD-Conv, trains end-to-end, and reports IoU/mIoU on the OpenOccupancy benchmark against published baselines. No equation in Sections 3.2-3.4 defines a predicted quantity in terms of the quantity being predicted; Eq. (4)-(5) merely sample a trained polar feature grid into Cartesian voxels using a coordinate transform and trilinear interpolation, with the final head and loss inherited from CONet [16], an external work. The only notable self-citation is [4] (PARTNER), co-authored by the first author, used as a design antecedent for decomposed attention (e.g., 'the output feature of the last decomposed attention [4]') and as an ablation comparator; this is not load-bearing because PVP's improvements are established by end-to-end training measured against the external benchmark, not by invoking PARTNER's results. Concerns about baseline protocol alignment in Section 4.1.2 ('All baseline implementations were aligned with the benchmark [16] to ensure fairness in comparison') are a validity and correctness risk about unrerun baselines, not a circularity reduction; the reported gains are empirical and could in principle be wrong, but they are not forced by construction. Score 1 reflects only a minor, non-load-bearing self-citation; the core result remains independently supported by external benchmark evaluation.

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

No ad hoc physical constants or fitted parameters in the classical sense; the model's weights are learned. The entries above reflect hand-chosen architectural hyperparameters and unstated design values that influence the reported performance, plus the benchmark and resampling assumptions the central claim depends on.

free parameters (3)
  • Window size Wg for local condense attention
    Chosen by hand, not reported numerically; controls how much the 3D volume is compressed before global attention.
  • Condensed feature downsampling factor S
    Appears in the condensed map resolution R/S x A/S x Z/S; the value is never stated, yet it determines the token count for global decomposed attention.
  • Stacking order of PD-Conv blocks = serial (1x3x3, 3x1x3, 3x3x1)
    The paper notes stacking order does not significantly affect performance and selects one arrangement; the choice is not derived.
assumptions (3)
  • domain assumption The OpenOccupancy benchmark annotations and evaluation protocol are accurate and are applied identically to all compared methods.
    All conclusions hinge on the validity of the benchmark, which is taken as a given.
  • domain assumption Polar-to-Cartesian trilinear sampling in Eq. (5) preserves enough information for accurate occupancy classification.
    The model predicts in polar space and resamples to Cartesian output; any information loss at this resampling would degrade the reported results.
  • ad hoc to paper Ablations run for 5 epochs are representative of the final 25-epoch results.
    All component-wise conclusions in Tables 2, 3, and 5 come from 5-epoch training, while final SOTA numbers use 25 epochs; the stability of the gains across this gap is not shown.

how reviews work

0 comments
Cite this review

Pith. "Pith review of PVP: Polar Representation Boost for 3D Semantic Occupancy Prediction." pith.science (2026). https://pith.science/paper/FZCEL3U2

@misc{pith2026241207616,
  author       = {Pith},
  title        = {Pith review of: PVP: Polar Representation Boost for 3D Semantic Occupancy Prediction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FZCEL3U2}},
  note         = {Machine review of arXiv:2412.07616}
}
read the original abstract

Recently, polar coordinate-based representations have shown promise for 3D perceptual tasks. Compared to Cartesian methods, polar grids provide a viable alternative, offering better detail preservation in nearby spaces while covering larger areas. However, they face feature distortion due to non-uniform division. To address these issues, we introduce the Polar Voxel Occupancy Predictor (PVP), a novel 3D multi-modal predictor that operates in polar coordinates. PVP features two key design elements to overcome distortion: a Global Represent Propagation (GRP) module that integrates global spatial data into 3D volumes, and a Plane Decomposed Convolution (PD-Conv) that simplifies 3D distortions into 2D convolutions. These innovations enable PVP to outperform existing methods, achieving significant improvements in mIoU and IoU metrics on the OpenOccupancy dataset.

Figures

Figures reproduced from arXiv: 2412.07616 by the authors.

Figure 1
Figure 1. An illustration of polar feature distortion: The non [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. The pipeline of PVP. Our proposed PVP consists of three components: 1) The grid-based feature extraction and fusion mod￾ule, which includes a 3D Voxel-based Backbone with PD-Conv for feature extraction and a 3D Image backbone for 2D to 3D feature conversion. 2) The GRP Module utilizes attention mechanisms to capture road structures from the scene volume and accurately propagate features to their correct locations. 3… view at source ↗
Figure 4
Figure 4. GRP module encompasses two types of attention sub-modules: 1) Local condense attention for condensing multi-modal local [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: A diagram illustrating PD-Conv. PD-Conv simplifies the [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 37 canonical work pages

  1. [16]

    Openoccupancy: A large scale benchmark for sur- rounding semantic occupancy perception

    Xiaofeng Wang, Zheng Zhu, Wenbo Xu, Yunpeng Zhang, Yi Wei, Xu Chi, Yun Ye, Dalong Du, Jiwen Lu, and Xingang Wang. Openoccupancy: A large scale benchmark for sur- rounding semantic occupancy perception. 2023. 1, 3, 4, 6, 7, 8

    work page 2023

  2. [1]

    Pointocc: Cylindrical tri-perspective view for point-based 3d semantic occupancy prediction

    Sicheng Zuo, Wenzhao Zheng, Yuanhui Huang, Jie Zhou, and Jiwen Lu. Pointocc: Cylindrical tri-perspective view for point-based 3d semantic occupancy prediction. arXiv preprint arXiv:2308.16896, 2023. 1, 5, 6, 7

    arXiv 2023

  3. [2]

    3d object detection with pointformer

    Xuran Pan, Zhuofan Xia, Shiji Song, Li Erran Li, and Gao Huang. 3d object detection with pointformer. In Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 7463–7472, 2021. 1

    work page 2021

  4. [3]

    Lif-seg: Lidar and camera image fusion for 3d lidar semantic segmentation

    Lin Zhao, Hui Zhou, Xinge Zhu, Xiao Song, Hongsheng Li, and Wenbing Tao. Lif-seg: Lidar and camera image fusion for 3d lidar semantic segmentation. IEEE Transactions on Multimedia, 2023. 1, 3

    work page 2023

  5. [4]

    Partner: Level up the polar representa- tion for lidar 3d object detection, 2023

    Ming Nie, Yujing Xue, Chunwei Wang, Chaoqiang Ye, Hang Xu, Xinge Zhu, Qingqiu Huang, Michael Bi Mi, Xinchao Wang, and Li Zhang. Partner: Level up the polar representa- tion for lidar 3d object detection, 2023. 1, 2, 3, 5, 8

    work page 2023

  6. [5]

    Deep learning-enabled 3d multimodal fusion of cone-beam ct and intraoral mesh scans for clinically applica- ble tooth-bone reconstruction

    Jiaxiang Liu, Jin Hao, Hangzheng Lin, Wei Pan, Jianfei Yang, Yang Feng, Gaoang Wang, Jin Li, Zuolin Jin, Zhihe Zhao, et al. Deep learning-enabled 3d multimodal fusion of cone-beam ct and intraoral mesh scans for clinically applica- ble tooth-bone reconstruction. Patterns, 4(9), 2023. 1

    work page 2023

  7. [6]

    Center- based 3d object detection and tracking

    Tianwei Yin, Xingyi Zhou, and Philipp Krahenbuhl. Center- based 3d object detection and tracking. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11784–11793, 2021. 1

    work page 2021

  8. [7]

    3d semantic segmentation with submani- fold sparse convolutional networks

    Benjamin Graham, Martin Engelcke, and Laurens Van Der Maaten. 3d semantic segmentation with submani- fold sparse convolutional networks. In Proceedings of the IEEE conference on computer vision and pattern recogni- tion, pages 9224–9232, 2018. 1

    work page 2018

Show all 43 references
  1. [8]

    Segcloud: Semantic segmen- tation of 3d point clouds

    Lyne Tchapmi, Christopher Choy, Iro Armeni, JunYoung Gwak, and Silvio Savarese. Segcloud: Semantic segmen- tation of 3d point clouds. In 2017 international conference on 3D vision (3DV), pages 537–547. IEEE, 2017. 1

    2017

  2. [9]

    Mseg3d: Multi-modal 3d semantic segmentation for autonomous driv- ing

    Jiale Li, Hang Dai, Hao Han, and Yong Ding. Mseg3d: Multi-modal 3d semantic segmentation for autonomous driv- ing. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , pages 21694–21704,

  3. [10]

    Hydro-3d: Hybrid object detection and tracking for co- operative perception using 3d lidar

    Zonglin Meng, Xin Xia, Runsheng Xu, Wei Liu, and Jiaqi Ma. Hydro-3d: Hybrid object detection and tracking for co- operative perception using 3d lidar. IEEE Transactions on Intelligent Vehicles, 2023. 1, 3

    2023

  4. [11]

    Supfusion: Supervised lidar- camera fusion for 3d object detection

    Yiran Qin, Chaoqun Wang, Zijian Kang, Ningning Ma, Zhen Li, and Ruimao Zhang. Supfusion: Supervised lidar- camera fusion for 3d object detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 22014–22024, 2023. 1, 3

    2023

  5. [12]

    Ai-enabled automatic multimodal fusion of cone-beam ct and intraoral scans for intelligent 3d tooth- bone reconstruction and clinical applications

    Jin Hao, Jiaxiang Liu, Jin Li, Wei Pan, Ruizhe Chen, Huimin Xiong, Kaiwei Sun, Hangzheng Lin, Wanlu Liu, Wanghui Ding, et al. Ai-enabled automatic multimodal fusion of cone-beam ct and intraoral scans for intelligent 3d tooth- bone reconstruction and clinical applications. arX...

    2022 arXiv

  6. [13]

    Refined individual tooth segmentation from cone beam ct images

    Tuo Zheng, Jiaxiang Liu, Yang Feng, Junhui Lv, and Zuozhu Liu. Refined individual tooth segmentation from cone beam ct images. In 2024 IEEE International Symposium on Biomedical Imaging (ISBI), pages 1–5. IEEE, 2024. 1

    2024

  7. [14]

    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. arXiv preprint arXiv:2302.07817, 2023. 1, 6, 7

    2023 arXiv

  8. [15]

    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. arXiv preprint arXiv:2304.05316 , 2023. 1

    2023 arXiv

  9. [17]

    Surroundocc: Multi-camera 3d oc- cupancy prediction for autonomous driving

    Yi Wei, Linqing Zhao, Wenzhao Zheng, Zheng Zhu, Jie Zhou, and Jiwen Lu. Surroundocc: Multi-camera 3d oc- cupancy prediction for autonomous driving. arXiv preprint arXiv:2303.09551, 2023. 1, 3

    2023 arXiv

  10. [18]

    Scalable geometric fracture assembly via co-creation space among assemblers

    Ruiyuan Zhang, Jiaxiang Liu, Zexi Li, Hao Dong, Jie Fu, and Chao Wu. Scalable geometric fracture assembly via co-creation space among assemblers. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 7269–7277, 2024. 1

    2024

  11. [19]

    Sa-convonet: Sign-agnostic optimization of convolutional occupancy networks

    Jiapeng Tang, Jiabao Lei, Dan Xu, Feiying Ma, Kui Jia, and Lei Zhang. Sa-convonet: Sign-agnostic optimization of convolutional occupancy networks. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 6504–6513, 2021. 1

    2021

  12. [20]

    Bevformer: Learning bird’s-eye-view representation from multi-camera images via spatiotemporal transformers, 2022

    Zhiqi Li, Wenhai Wang, Hongyang Li, Enze Xie, Chong- hao Sima, Tong Lu, Qiao Yu, and Jifeng Dai. Bevformer: Learning bird’s-eye-view representation from multi-camera images via spatiotemporal transformers, 2022. 1, 3, 4

    2022

  13. [21]

    Feature pyra- mid networks for object detection

    Tsung-Yi Lin, Piotr Doll ´ar, Ross Girshick, Kaiming He, Bharath Hariharan, and Serge Belongie. Feature pyra- mid networks for object detection. In Proceedings of the IEEE conference on computer vision and pattern recogni- tion, pages 2117–2125, 2017. 1, 4

    2017

  14. [22]

    Polarformer: Multi- camera 3d object detection with polar transformer

    Yanqin Jiang, Li Zhang, Zhenwei Miao, Xiatian Zhu, Jin Gao, Weiming Hu, and Yu-Gang Jiang. Polarformer: Multi- camera 3d object detection with polar transformer. In Pro- ceedings of the AAAI Conference on Artificial Intelligence , volume 37, pages 1042–1050, 2023. 2

    2023

  15. [23]

    Craft: Camera-radar 3d object detection with spatio-contextual fusion transformer

    Youngseok Kim, Sanmin Kim, Jun Won Choi, and Dong- suk Kum. Craft: Camera-radar 3d object detection with spatio-contextual fusion transformer. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, pages 1160–1168, 2023. 2

    2023

  16. [24]

    Semantic seg- mentation of 3d lidar data using deep learning: a re- view of projection-based methods

    Alok Jhaldiyal and Navendu Chaudhary. Semantic seg- mentation of 3d lidar data using deep learning: a re- view of projection-based methods. Applied Intelligence , 53(6):6844–6855, 2023. 2 9

    2023

  17. [25]

    Polarpoint-bev: Bird-eye- view perception in polar points for explainable end-to-end autonomous driving

    Yuchao Feng and Yuxiang Sun. Polarpoint-bev: Bird-eye- view perception in polar points for explainable end-to-end autonomous driving. IEEE Transactions on Intelligent Vehi- cles, 2024. 2

    2024

  18. [26]

    Polarnet: An improved grid representation for online lidar point clouds se- mantic segmentation

    Yang Zhang, Zixiang Zhou, Philip David, Xiangyu Yue, Ze- rong Xi, Boqing Gong, and Hassan Foroosh. Polarnet: An improved grid representation for online lidar point clouds se- mantic segmentation. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recogn...

    2020

  19. [27]

    Cylindrical and asymmetrical 3d convolution networks for lidar segmenta- tion

    Xinge Zhu, Hui Zhou, Tai Wang, Fangzhou Hong, Yuexin Ma, Wei Li, Hongsheng Li, and Dahua Lin. Cylindrical and asymmetrical 3d convolution networks for lidar segmenta- tion. In Proceedings of the IEEE/CVF conference on com- puter vision and pattern recognition, 2021. 2, 3, 5, 8

    2021

  20. [28]

    Polarstream: Streaming object detection and segmentation with polar pil- lars

    Qi Chen, Sourabh V ora, and Oscar Beijbom. Polarstream: Streaming object detection and segmentation with polar pil- lars. Advances in Neural Information Processing Systems ,

  21. [29]

    One training for multiple deployments: Polar-based adap- tive bev perception for autonomous driving

    Huitong Yang, Xuyang Bai, Xinge Zhu, and Yuexin Ma. One training for multiple deployments: Polar-based adap- tive bev perception for autonomous driving. arXiv preprint arXiv:2304.00525, 2023. 3

    2023 arXiv

  22. [30]

    Se- mantickitti: A dataset for semantic scene understanding of lidar sequences, 2019

    Jens Behley, Martin Garbade, Andres Milioto, Jan Quen- zel, Sven Behnke, Cyrill Stachniss, and Juergen Gall. Se- mantickitti: A dataset for semantic scene understanding of lidar sequences, 2019. 3

    2019

  23. [31]

    Lmscnet: Lightweight multiscale 3d semantic completion,

    Luis Rold ˜ao, Raoul de Charette, and Anne Verroust-Blondet. Lmscnet: Lightweight multiscale 3d semantic completion,

  24. [32]

    Sparse single sweep lidar point cloud segmentation via learning contextual shape priors from scene completion, 2021

    Xu Yan, Jiantao Gao, Jie Li, Ruimao Zhang, Zhen Li, Rui Huang, and Shuguang Cui. Sparse single sweep lidar point cloud segmentation via learning contextual shape priors from scene completion, 2021. 3, 6, 7

    2021

  25. [33]

    Anisotropic convolutional networks for 3d semantic scene completion, 2020

    Jie Li, Kai Han, Peng Wang, Yu Liu, and Xia Yuan. Anisotropic convolutional networks for 3d semantic scene completion, 2020. 3, 6, 7

    2020

  26. [34]

    Lift, splat, shoot: Encoding images from arbitrary camera rigs by implicitly unprojecting to 3d, 2020

    Jonah Philion and Sanja Fidler. Lift, splat, shoot: Encoding images from arbitrary camera rigs by implicitly unprojecting to 3d, 2020. 4

    2020

  27. [35]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceed- ings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016. 4

    2016

  28. [36]

    Bevfusion: Multi- task multi-sensor fusion with unified bird’s-eye view repre- sentation

    Zhijian Liu, Haotian Tang, Alexander Amini, Xinyu Yang, Huizi Mao, Daniela L Rus, and Song Han. Bevfusion: Multi- task multi-sensor fusion with unified bird’s-eye view repre- sentation. In 2023 IEEE international conference on robotics and automation (ICRA), pages 2774–2781. I...

    2023

  29. [37]

    Scf-net: Learning spatial contextual fea- tures for large-scale point cloud segmentation

    Siqi Fan, Qiulei Dong, Fenghua Zhu, Yisheng Lv, Peijun Ye, and Fei-Yue Wang. Scf-net: Learning spatial contextual fea- tures for large-scale point cloud segmentation. In Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 14504–14513, 2021. 5

    2021

  30. [38]

    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 Conference on Computer Vision and Pattern R...

    2020

  31. [39]

    Monoscene: Monoc- ular 3d semantic scene completion, 2022

    Anh-Quan Cao and Raoul de Charette. Monoscene: Monoc- ular 3d semantic scene completion, 2022. 6, 7

    2022

  32. [40]

    3d sketch-aware semantic scene completion via semi-supervised structure prior, 2020

    Xiaokang Chen, Kwan-Yee Lin, Chen Qian, Gang Zeng, and Hongsheng Li. 3d sketch-aware semantic scene completion via semi-supervised structure prior, 2020. 6, 7

    2020

  33. [41]

    Toothsegnet: image degradation meets tooth segmentation in cbct images

    Jiaxiang Liu, Tianxiang Hu, Yang Feng, Wanghui Ding, and Zuozhu Liu. Toothsegnet: image degradation meets tooth segmentation in cbct images. In 2023 IEEE 20th Interna- tional Symposium on Biomedical Imaging (ISBI), pages 1–5. IEEE, 2023. 6

    2023

  34. [42]

    Daquan Zhou, Zhiding Yu, Enze Xie, Chaowei Xiao, An- imashree Anandkumar, Jiashi Feng, and Jose M. Alvarez. Understanding the robustness in vision transformers. In Ka- malika Chaudhuri, Stefanie Jegelka, Le Song, Csaba Szepes- vari, Gang Niu, and Sivan Sabato, editors, Proceed...

    2022

  35. [43]

    Swin transformer: Hierarchical vision transformer using shifted windows

    Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, and Baining Guo. Swin transformer: Hierarchical vision transformer using shifted windows. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021. 8 10

    2021

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.