Pith. sign in

REVIEW 3 major objections 7 minor 34 references

4D-ROLLS: 4D Radar Occupancy Learning via LiDAR Supervision

T0 review · 3 major / 7 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read 4D-ROLLS claims that a radar-only network can estimate 3D occupancy at near-LiDAR accuracy, trained with no manual labels by using the LiDAR point cloud purely as a training-time teacher.

desk verdict A useful and clearly presented weakly supervised radar occupancy pipeline; the novelty holds up, but the evaluation is partly circular and the all-weather claim is thinner than the abstract suggests. read the letter →

arxiv 2505.13905 v1 pith:J5JZCEU4 submitted 2025-05-20 cs.CV cs.RO

classification cs.CVcs.RO
keywords 4DradaroccupancyestimationweaklysupervisedlearningLiDARsupervisionpseudo-labelgenerationtri-perspectiveviewautonomousdrivingall-weatherperception
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 tries to establish that 4D radar point clouds alone are enough to produce dense 3D occupancy maps that rival LiDAR-based ones, provided the radar network is trained under LiDAR's supervision. Because LiDAR and cameras fail in rain, snow, fog, and smoke while 4D radar keeps sensing, a radar-only occupancy model would let an autonomous vehicle keep a reliable picture of occupied and free space in exactly the conditions where current sensors degrade. The paper's move is to turn an unlabeled LiDAR scan into two free supervisory signals: occupied and free query points sampled along each LiDAR ray, and a per-column height map. The radar network is trained on those signals, then fine-tuned against a self-supervised LiDAR occupancy model to prune spurious detections. The reported result is that this radar model beats LiDAR-derived occupancy baselines re-targeted to radar input, transfers across datasets, and runs at about 30 Hz.

What carries the argument

The load-bearing object is the ray-sampled occupancy query pair. For a LiDAR point $p_i$ and unit ray direction $\mathbf{u}$ from the sensor to the point, the occupied query $R^+_i = p_i + r\mathbf{u}$ and the free query $R^-_i = p_i - r\mathbf{u}$ encode the assumption that the sensor-to-surface ray is empty and the cell just beyond the surface is filled; the whole training signal is a set of such pairs. The second mechanism is the LiDAR height map, a projection in which each XY grid cell stores the maximum Z value of the LiDAR points above it, supervised at two network depths by a masked MSE loss computed only where radar points exist. These two pseudo-label types drive a sparse tri-perspective-view encoder (three perpendicular 2D planes: bird's-eye, front, side) feeding a dense UNO-style decoder and an occupancy head, with total loss $L = \omega_1 L^1_{\mathrm{height}} + \omega_2 L^2_{\mathrm{height}} + \omega_3 L_{\mathrm{occ}}$, followed by stage-2 fine-tuning against a LiDAR self-supervised occupancy model.

What would settle it

Build a small labeled evaluation set of scenes with objects thinner than the occupancy voxel (0.4 m) or with surfaces radar can penetrate, such as foliage, wire mesh, or thin plastic panels, and compare the trained model's occupancy map against a dense ground-truth scan. If the model marks the space behind these surfaces as free even where genuine radar returns exist behind them, the LiDAR ray-casting assumption has propagated into the predictions; measuring a systematic occupancy error on such objects that grows with their transparency would settle whether the supervision bias is real.

Watch

Extended reading notes

Core claim

The central claim, stated as the authors would state it, is that the gap between sparse, noisy 4D radar and LiDAR-grade occupancy is mostly a supervision gap, not a sensing gap. For each LiDAR point $p_i$ the method labels the sample $R^+_i = p_i + r\mathbf{u}$ (just behind the surface) as occupied and $R^-_i = p_i - r\mathbf{u}$ (toward the sensor) as free, converting a raw LiDAR scan into a dense set of occupancy queries for free. A second signal, the LiDAR height map recording the maximum Z value in each XY voxel, is used to pin down the vertical axis where radar's height resolution is weakest. A two-stage procedure then trains the radar network: stage 1 fits the occupancy field and height maps to the pseudo-labels, and stage 2 fine-tunes it against a self-supervised LiDAR occupancy model, removing false occupied regions that radar's sparse coverage produces. The authors claim this yields radar-only occupancy estimation that significantly outperforms the LiDAR-occupancy baselines adapted to radar input on all key metrics, remains stable in fog and smoke where LiDAR supervision itself degrades, and generalizes across datasets; they also acknowledge the trade-off that radar returns from behind objects, which LiDAR supervision ignores, can bias the learned model.

Load-bearing premise

The whole training signal rests on assuming that each LiDAR point lies exactly on an object surface, so every voxel between sensor and point is empty and the voxel just behind the point is occupied; if the object is thinner than the offset $r$, partially hidden, or mis-measured by LiDAR, the pseudo-labels themselves are wrong and the network faithfully learns those errors.

Editorial extensions

If this is right

  • Autonomous vehicles could carry radar as the sole occupancy sensor at inference time, remaining functional in fog, smoke, rain, and snow where LiDAR and cameras degrade.
  • Any LiDAR-equipped vehicle produces free, unlimited training labels for radar occupancy, so the method scales with data collection rather than annotation budgets.
  • The stage-2 fine-tuning recipe — train a denser sensor's self-supervised occupancy model and use it to prune a sparser sensor's false predictions — transfers geometric fidelity across sensor types without manual labels.
  • The learned occupancy representation serves as a pretrained backbone for BEV segmentation and point-cloud occupancy prediction with only simple fine-tuning, indicating it captures general scene structure.
  • At about 33 ms per frame on a mid-range GPU, the lightweight network meets real-time constraints for robotic navigation and planning.

Reading between the lines

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

  • If the supervision signal is the bottleneck, then upgrading the teacher — a denser LiDAR, or LiDAR fused with cameras — should directly raise radar occupancy accuracy in normal weather; this is a testable consequence the paper does not run.
  • The height-map branch specifically compensates radar's weak vertical resolution; ablating it would quantify how much of the reported gain is Z-axis correction versus occupancy-query learning.
  • Radar returns that persist behind LiDAR-defined surfaces are treated as errors to be pruned, but they could instead be harvested as evidence of transparent or thin objects, turning the known penetration bias into a material-boundary signal.
  • The reported cross-dataset tilt from the NTU sensor mount suggests that applying the method to a new radar first requires precise extrinsic calibration; a small misalignment could otherwise be baked into the learned occupancy field.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 7 minor

Summary. The paper proposes 4D-ROLLS, a 4D radar occupancy estimation method trained with weak supervision from LiDAR. The method generates pseudo-labels consisting of occupancy queries (occupied and free points placed along LiDAR rays) and a LiDAR height map, and uses these in a multi-term loss (Eq. 3) with a sparse TPV encoder and a dense decoder. A second fine-tuning stage aligns the radar model with a self-supervised LiDAR occupancy model. Experiments on the MSC and NTU datasets compare against LiDAR occupancy methods adapted to radar input, reporting lower Chamfer distance, L2 error, and relative L2 error, qualitative robustness under smoke, cross-dataset generalization, transfer to BEV segmentation and point cloud occupancy prediction, and inference at about 30 Hz on a 4060 GPU.

Significance. If validated, the paper addresses a practically important gap: 4D radar point clouds are sparser and noisier than LiDAR, and no open-source 4D radar point-cloud occupancy method exists. The proposed training recipe is simple, the network is lightweight, and the authors provide code. The cross-dataset test and downstream-task transfer are useful evidence of representation quality. However, the central quantitative evaluation is weakened by the fact that the LiDAR-derived pseudo-labels are used both as training targets and as evaluation reference, so the reported gains may measure imitation of LiDAR ray casting rather than independent occupancy accuracy. The paper's all-weather claim is also not quantitatively supported, and the authors themselves concede a bias in the learned world model for radar-penetrated objects. The contribution is valuable as a LiDAR-supervised distillation approach, but the claims need to be reframed or supported by independent ground truth.

major comments (3)
  1. [Section III.A and Tables I/II] The occupancy queries R+ and R- are generated by ray casting from LiDAR points, and these same queries are used as the training target L_occ in Eq. (3) and as the reference for the CD, NFCD, AR, and L2 metrics in Tables I and II. Consequently, the reported gains of Ours-stage1 and Ours-stage2 measure how well the radar network reproduces the LiDAR-derived pseudo-labels, not occupancy accuracy against independent ground truth. This is especially problematic for the all-weather claim: in fog or smoke the LiDAR pseudo-labels are unreliable, and in clear weather radar can detect objects that the ray-casting assumption marks as free. The paper should either add an evaluation with independent occupancy ground truth (e.g., manually labeled voxels or geometric annotations not derived from the same LiDAR scans) or substantially soften the claims of all-weather occupancy accuracy.
  2. [Section IV.B, last paragraph] The text concedes that radar may capture objects after penetration while the LiDAR supervision signal completely ignores them, 'causing the network to learn a biased world model.' This admission directly qualifies the central claim that the method preserves all-weather sensing. The only degraded-weather result is qualitative (Fig. 4(a)), where LiDAR cannot provide a reference; thus the quantitative tables do not support the abstract's assertion of robustness in degraded environments. Please either provide quantitative evaluation in degraded conditions with independent labels or reframe the contribution as LiDAR-style occupancy distillation rather than all-weather occupancy estimation.
  3. [Section IV.B and Section IV.A] The baseline adaptation protocol is underspecified. It is not stated whether ALSO-R and UNO-R are trained with the same LiDAR-generated occupancy queries and height maps as 4D-ROLLS, what loss functions are used, or how hyperparameters are chosen. Also, the LiDAR-input methods (ALSO-L, UNO-L) are labeled as references but are not upper bounds for a radar-only system; please clarify their role. A fair comparison requires identical supervision and evaluation protocol for all radar-input models.
minor comments (7)
  1. [Section IV.A] The text contains typos: 'Jeston orin' should be 'Jetson Orin' and 'repectively' should be 'respectively'; these appear in Section IV.A and in the footnotes of Tables I and II.
  2. [Section IV.A and Table I] Section IV.A lists NFCD as a metric, but Table I reports only CD, AR, and L2; either add NFCD to Table I or remove it from the metric list for consistency.
  3. [Section III.A] The equations defining R+ and R- are not numbered; number them for clarity and ease of reference.
  4. [Section IV.A] The term 'relative L2 error (AR)' is not standard and the abbreviation AR is not defined; please rename or define it explicitly.
  5. [Fig. 4 caption] The caption uses '(cross-dataset training)' twice and does not specify which panels correspond to same-dataset versus cross-dataset testing; please annotate the panels clearly.
  6. [References] Reference [1] appears to cite 'COLMAP' but the title describes an occupancy grid mapping framework; please verify the citation and correct the author or title if needed.
  7. [Section IV.A] The statement that the MSC dataset comprises 90,000 frames is not connected to the per-sequence results in Table I; clarify whether training uses all frames or only the listed sequences.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: LiDAR pseudo-labels are an external supervisory source, and the all-weather claim is explicitly qualitative, not a fitted prediction.

full rationale

4D-ROLLS does not derive its occupancy output from its own inputs. The pseudo-label generator in Sec. III.A converts external LiDAR points into R+/R- occupancy queries and a LiDAR height map, and the network is trained with L = w1*L1_height + w2*L2_height + w3*L_occ (Eq. 3), where L_occ compares the predicted occupancy map to those LiDAR-derived queries. The evaluation metrics (CD, NFCD, L2, AR) are applied consistently to all methods against a LiDAR-based reference, so the reported comparison is a fair test of how well each radar model imitates LiDAR occupancy, not a derivation from fitted constants or from the radar input itself. The paper explicitly bounds the all-weather claim: 'Since LiDAR degrades in smokes and becomes unreliable as ground truth, we can only qualitatively demonstrate the stability of our method' (Sec. IV.B), and it concedes 'in extreme cases, radar may capture objects after penetration, while the LiDAR supervision signal completely ignores them, causing the network to learn a biased world model.' These are acknowledged external-validity limitations, not circular reductions. Self-citations to the authors' earlier radar odometry and segmentation work ([4], [5]) are contextual and do not carry the central claim. No step in the claimed derivation chain reduces to its own input by construction.

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

The method rests on modeling choices rather than physical postulates: ray-casting label generation, a height-map proxy, grid resolution and range choices, and unstated loss weights. No new physical entities are introduced.

free parameters (5)
  • Occupancy query offset r = not reported
    Controls the placement of R+ and R- sampling points along each LiDAR ray in Section III.A. Directly determines which voxels are labeled occupied or free, and its value is never specified.
  • Occupancy voxel size = 0.4 m x 0.4 m x 0.4 m
    Chosen grid resolution for occupancy estimation in Section IV.A. Affects label density, memory use, and metric values.
  • Estimation range = (0, 51.2) m along X, (-25.6, 25.6) m along Y, (-3, 3) m along Z
    Trims the field of view in Section IV.A. Affects difficulty and comparability of reported metrics.
  • Loss weights omega_1, omega_2, omega_3 = not reported
    Balance L1_height, L2_height, and L_occ in Eq. 3 of Section III.B. Values are not reported, so the relative importance of height supervision versus occupancy supervision is unspecified.
  • LiDAR height map XY grid resolution = not reported
    Determines which LiDAR maximum Z value is assigned to each radar-occupied XY voxel in Section III.A. The grid spacing is not stated.
assumptions (5)
  • domain assumption LiDAR ray assumption: all voxels along the ray from the sensor to a LiDAR point are unoccupied.
    Foundation of occupancy query generation in Section III.A. If multipath, occlusion, or LiDAR noise violates this, the free-space labels are wrong.
  • domain assumption Observed objects are thicker than the sampling offset r, so R+ lies inside the object surface.
    Section III.A states R+ is occupied 'as long as the object observed is not thinner than r'.
  • domain assumption The maximum LiDAR Z value in each XY voxel is a valid proxy for the occupancy height the radar should predict.
    Basis of LiDAR height map supervision in Section III.A. The radar's own vertical distribution is too discrete and noisy to supervise directly.
  • domain assumption LiDAR-radar extrinsic calibration is accurate enough to project radar points into LiDAR coordinates.
    Implied in Section III.A by the projection step. The authors themselves note an installation tilt in the NTU dataset, and misalignment would corrupt supervision.
  • domain assumption LiDAR-derived occupancy is a valid ground truth for quantitative evaluation.
    Tables I and II use LiDAR occupancy (from ALSO and UNO) to compute CD, NFCD, AR, and L2. This assumption fails in degraded weather and for radar-penetrable objects, as the authors acknowledge in Section IV.B.

how reviews work

0 comments
Cite this review

Pith. "Pith review of 4D-ROLLS: 4D Radar Occupancy Learning via LiDAR Supervision." pith.science (2026). https://pith.science/paper/J5JZCEU4

@misc{pith2026250513905,
  author       = {Pith},
  title        = {Pith review of: 4D-ROLLS: 4D Radar Occupancy Learning via LiDAR Supervision},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/J5JZCEU4}},
  note         = {Machine review of arXiv:2505.13905}
}
read the original abstract

A comprehensive understanding of 3D scenes is essential for autonomous vehicles (AVs), and among various perception tasks, occupancy estimation plays a central role by providing a general representation of drivable and occupied space. However, most existing occupancy estimation methods rely on LiDAR or cameras, which perform poorly in degraded environments such as smoke, rain, snow, and fog. In this paper, we propose 4D-ROLLS, the first weakly supervised occupancy estimation method for 4D radar using the LiDAR point cloud as the supervisory signal. Specifically, we introduce a method for generating pseudo-LiDAR labels, including occupancy queries and LiDAR height maps, as multi-stage supervision to train the 4D radar occupancy estimation model. Then the model is aligned with the occupancy map produced by LiDAR, fine-tuning its accuracy in occupancy estimation. Extensive comparative experiments validate the exceptional performance of 4D-ROLLS. Its robustness in degraded environments and effectiveness in cross-dataset training are qualitatively demonstrated. The model is also seamlessly transferred to downstream tasks BEV segmentation and point cloud occupancy prediction, highlighting its potential for broader applications. The lightweight network enables 4D-ROLLS model to achieve fast inference speeds at about 30 Hz on a 4060 GPU. The code of 4D-ROLLS will be made available at https://github.com/CLASS-Lab/4D-ROLLS.

Figures

Figures reproduced from arXiv: 2505.13905 by the authors.

Figure 1
Figure 1. We compare our method with the classic LiDAR-based [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overview of the Framework: including Occupancy Estimation, Prediction Results, and BEV Segmentation Performance [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Qualitative comparison on the MSC dataset. Despite the sparse 4D radar point cloud, our approach effectively infers [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Qualitative comparison on NTU dataset, including cross-dataset testing and intra-dataset testing. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: The results of downstream tasks using our method. [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

34 extracted references · 18 canonical work pages

  1. [1]

    Colmap: A memory-efficient occupancy grid mapping frame- work,

    A. Fisher, R. Cannizzaro, M. Cochrane, C. Nagahawatte, and J. L. Palmer, “Colmap: A memory-efficient occupancy grid mapping frame- work,”Robotics and Autonomous Systems, vol. 142, p. 103755, 2021

  2. [2]

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

    Y . Huang, W. Zheng, Y . Zhang, J. Zhou, and J. Lu, “Tri-perspective view for vision-based 3d semantic occupancy prediction,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2023, pp. 9223–9232

  3. [3]

    MetaOcc: Spatio-Temporal Fusion of Surround-View 4D Radar and Camera for 3D Occupancy Prediction with Dual Training Strategies

    L. Yang, L. Zheng, W. Ai, M. Liu, S. Li, Q. Lin, S. Yan, J. Bai, Z. Ma, and X. Zhu, “Metaocc: Surround-view 4d radar and camera fusion framework for 3d occupancy prediction with dual training strategies,” arXiv preprint arXiv:2501.15384, 2025

  4. [4]

    Efear-4d: Ego-velocity filtering for efficient and accurate 4d radar odometry,

    X. Wu, Y . Chen, Z. Li, Z. Hong, and L. Hu, “Efear-4d: Ego-velocity filtering for efficient and accurate 4d radar odometry,”IEEE Robotics and Automation Letters, vol. 9, no. 11, pp. 9828–9835, 2024

  5. [5]

    Get it for free: Radar segmentation without expert labels and its application in odometry and localization,

    S. Li, Z. Hong, Y . Chen, L. Hu, and J. Qin, “Get it for free: Radar segmentation without expert labels and its application in odometry and localization,”IEEE Robotics and Automation Letters, vol. 10, no. 3, pp. 2678–2685, 2025

  6. [6]

    4DRVO-Net: Deep 4D radar–visual odometry using multi-modal and multi-scale adaptive fusion,

    G. Zhuo, S. Lu, L. Xiong, H. Zhouins, L. Zheng, and M. Zhou, “4DRVO-Net: Deep 4D radar–visual odometry using multi-modal and multi-scale adaptive fusion,”IEEE Transactions on Intelligent Vehicles, 2023

  7. [7]

    Gaussian radar transformer for semantic segmentation in noisy radar data,

    M. Zeller, J. Behley, M. Heidingsfeld, and C. Stachniss, “Gaussian radar transformer for semantic segmentation in noisy radar data,”IEEE Robotics and Automation Letters, vol. 8, no. 1, pp. 344–351, 2022

  8. [8]

    Point cloud forecasting as a proxy for 4d occupancy forecasting,

    T. Khurana, P. Hu, D. Held, and D. Ramanan, “Point cloud forecasting as a proxy for 4d occupancy forecasting,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 1116–1124

Show all 34 references
  1. [9]

    Also: Automotive lidar self-supervision by occupancy estimation,

    A. Boulch, C. Sautier, B. Michele, G. Puy, and R. Marlet, “Also: Automotive lidar self-supervision by occupancy estimation,” inPro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 13 455–13 465

  2. [10]

    Multi- class road user detection with 3+ 1d radar in the view-of-delft dataset,

    A. Palffy, E. Pool, S. Baratam, J. F. Kooij, and D. M. Gavrila, “Multi- class road user detection with 3+ 1d radar in the view-of-delft dataset,” IEEE Robotics and Automation Letters, vol. 7, no. 2, pp. 4961–4968, 2022

  3. [11]

    Interfusion: Interaction-based 4d radar and lidar fusion for 3d object detection,

    L. Wang, X. Zhang, B. Xv, J. Zhang, R. Fu, X. Wang, L. Zhu, H. Ren, P. Lu, J. Li,et al., “Interfusion: Interaction-based 4d radar and lidar fusion for 3d object detection,” in2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2022, pp. 12 247–12 253

  4. [12]

    Dart: Implicit doppler tomography for radar novel view synthesis,

    T. Huang, J. Miller, A. Prabhakara, T. Jin, T. Laroia, Z. Kolter, and A. Rowe, “Dart: Implicit doppler tomography for radar novel view synthesis,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 24 118–24 129

  5. [13]

    4d iRIOM: 4D imaging radar inertial odometry and mapping,

    Y . Zhuang, B. Wang, J. Huai, and M. Li, “4d iRIOM: 4D imaging radar inertial odometry and mapping,”IEEE Robotics and Automation Letters, vol. 8, no. 6, pp. 3246–3253, 2023

  6. [14]

    Radarocc: Robust 3d occupancy prediction with 4d imaging radar,

    F. Ding, X. Wen, Y . Zhu, Y . Li, and C. X. Lu, “Radarocc: Robust 3d occupancy prediction with 4d imaging radar,”arXiv preprint arXiv:2405.14014, 2024

  7. [15]

    Dynamic occupancy grids for object detection: A radar-centric approach,

    M. P. Ronecker, M. Schratter, L. Kuschnig, and D. Watzenig, “Dynamic occupancy grids for object detection: A radar-centric approach,”arXiv preprint arXiv:2402.01488, 2024

  8. [16]

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

    S. Zuo, W. Zheng, Y . Huang, J. Zhou, and J. Lu, “Pointocc: Cylindrical tri-perspective view for point-based 3d semantic occupancy prediction,” arXiv preprint arXiv:2308.16896, 2023

  9. [17]

    Surroundocc: Multi-camera 3d occupancy prediction for autonomous driving,

    Y . Wei, L. Zhao, W. Zheng, Z. Zhu, J. Zhou, and J. Lu, “Surroundocc: Multi-camera 3d occupancy prediction for autonomous driving,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 21 729–21 740

  10. [18]

    Geocc: Geometrically enhanced 3d occupancy network with implicit-explicit depth fusion and contextual self-supervision,

    X. Tan, W. Wu, Z. Zhang, C. Fan, Y . Peng, Z. Zhang, Y . Xie, and L. Ma, “Geocc: Geometrically enhanced 3d occupancy network with implicit-explicit depth fusion and contextual self-supervision,”arXiv preprint arXiv:2405.10591, 2024

  11. [19]

    Occfusion: Multi- sensor fusion framework for 3d semantic occupancy prediction,

    Z. Ming, J. S. Berrio, M. Shan, and S. Worrall, “Occfusion: Multi- sensor fusion framework for 3d semantic occupancy prediction,”IEEE Transactions on Intelligent Vehicles, 2024

  12. [20]

    Licrocc: Teach radar for accurate semantic occupancy prediction using lidar and camera,

    Y . Ma, J. Mei, X. Yang, L. Wen, W. Xu, J. Zhang, X. Zuo, B. Shi, and Y . Liu, “Licrocc: Teach radar for accurate semantic occupancy prediction using lidar and camera,”IEEE Robotics and Automation Letters, 2024

  13. [21]

    Multi-class road user detection with 3+1d radar in the view-of-delft dataset,

    A. Palffy, E. Pool, S. Baratam, J. F. P. Kooij, and D. M. Gavrila, “Multi-class road user detection with 3+1d radar in the view-of-delft dataset,”IEEE Robotics and Automation Letters, vol. 7, no. 2, pp. 4961–4968, 2022

  14. [22]

    Ntu4dradlm: 4d radar-centric multi-modal dataset for localization and mapping,

    J. Zhang, H. Zhuge, Y . Liu, G. Peng, Z. Wu, H. Zhang, Q. Lyu, H. Li, C. Zhao, D. Kircali,et al., “Ntu4dradlm: 4d radar-centric multi-modal dataset for localization and mapping,” in2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC). IEEE, 2023...

  15. [23]

    Msc-rad4r: Ros-based automotive dataset with 4d radar,

    M. Choi, S. Yang, S. Han, Y . Lee, M. Lee, K. H. Choi, and K.-S. Kim, “Msc-rad4r: Ros-based automotive dataset with 4d radar,”IEEE Robotics and Automation Letters, 2023

  16. [24]

    Up-to-down network: Fusing multi-scale context for 3d semantic scene completion,

    H. Zou, X. Yang, T. Huang, C. Zhang, Y . Liu, W. Li, F. Wen, and H. Zhang, “Up-to-down network: Fusing multi-scale context for 3d semantic scene completion,” in2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2021, pp. 16–23

  17. [25]

    Openoccupancy: A large scale benchmark for surrounding semantic occupancy perception,

    X. Wang, Z. Zhu, W. Xu, Y . Zhang, Y . Wei, X. Chi, Y . Ye, D. Du, J. Lu, and X. Wang, “Openoccupancy: A large scale benchmark for surrounding semantic occupancy perception,” inProceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 17 850–17 859

  18. [26]

    Uno: Unsupervised occupancy fields for perception and forecasting,

    B. Agro, Q. Sykora, S. Casas, T. Gilles, and R. Urtasun, “Uno: Unsupervised occupancy fields for perception and forecasting,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 14 487–14 496

  19. [27]

    Scan context: Egocentric spatial descriptor for place recognition within 3d point cloud map,

    G. Kim and A. Kim, “Scan context: Egocentric spatial descriptor for place recognition within 3d point cloud map,” in2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2018, pp. 4802–4809

  20. [28]

    V oxelnet: End-to-end learning for point cloud based 3d object detection,

    Y . Zhou and O. Tuzel, “V oxelnet: End-to-end learning for point cloud based 3d object detection,” inProceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 4490–4499

  21. [29]

    Bevformer: learning bird’s-eye-view representation from lidar- camera via spatiotemporal transformers,

    Z. Li, W. Wang, H. Li, E. Xie, C. Sima, T. Lu, Q. Yu, and J. Dai, “Bevformer: learning bird’s-eye-view representation from lidar- camera via spatiotemporal transformers,”IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024

  22. [30]

    U-net: Convolutional net- works for biomedical image segmentation,

    O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional net- works for biomedical image segmentation,” inMedical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18. ...

  23. [31]

    Are we ready for autonomous driving? the kitti vision benchmark suite,

    A. Geiger, P. Lenz, and R. Urtasun, “Are we ready for autonomous driving? the kitti vision benchmark suite,” in2012 IEEE conference on computer vision and pattern recognition. IEEE, 2012, pp. 3354–3361

  24. [32]

    K-radar: 4d radar object detection for autonomous driving in various weather conditions,

    D.-H. Paek, S.-H. Kong, and K. T. Wijaya, “K-radar: 4d radar object detection for autonomous driving in various weather conditions,” Advances in Neural Information Processing Systems, vol. 35, pp. 3819– 3829, 2022

  25. [33]

    Deep high-resolution representation learning for human pose estimation,

    K. Sun, B. Xiao, D. Liu, and J. Wang, “Deep high-resolution representation learning for human pose estimation,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019, pp. 5693–5703

  26. [34]

    Semantickitti: A dataset for semantic scene understanding of lidar sequences,

    J. Behley, M. Garbade, A. Milioto, J. Quenzel, S. Behnke, C. Stachniss, and J. Gall, “Semantickitti: A dataset for semantic scene understanding of lidar sequences,” inProceedings of the IEEE/CVF international conference on computer vision, 2019, pp. 9297–9307

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.