Pith. sign in

REVIEW 4 major objections 5 minor 53 references

Uncertainty-aware Accurate Elevation Modeling for Off-road Navigation via Neural Processes

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Neural-process terrain models keep sharp elevation changes and uncertainty without sacrificing accuracy.

desk verdict A plausible new integration of semantic-conditioned neural processes for terrain elevation modeling, with solid results but an under-specified evaluation protocol that must be fixed before the accuracy claims are fully trustworthy. read the letter →

arxiv 2508.03890 v2 pith:E4YLA7WP submitted 2025-08-05 cs.RO

Sanghun Jung , Daehoon Gwak , Byron Boots , James Hays This is my paper · ORCID
classification cs.RO
keywords terrainelevationmodelingoff-roadnavigationneuralprocessespredictiveuncertaintyball-queryattentionbird's-eye-viewgridLiDAR-camerafusiontemporalaggregation
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

Terrain elevation maps for off-road robots must be accurate enough to expose ditches and cliffs and honest enough to say where they are unsure, but existing Gaussian-process and neural methods either run too slowly, smooth away sharp edges, or degrade accuracy when trained to output uncertainty. This paper proposes a Neural Process model that is conditioned on semantic features from LiDAR and cameras and uses local ball-query attention, so it can estimate sharp elevation changes and a predictive distribution at the same time. On three off-road datasets collected from grassy hills, desert, and dirt trails with negative obstacles, the method reports the lowest total elevation and slope errors among the compared baselines and curvature improvements over the neural baselines, while also keeping accuracy drops small when transferred to unseen terrain. The practical payoff is that a planner can trust both the height and the uncertainty of every cell in a bird's-eye-view grid.

What carries the argument

The central object is a semantic-conditioned Attentive Neural Process: a Neural Process in which a global latent variable $z$ and a deterministic representation $r_C$, computed from context points, condition a decoder that outputs a factorized Gaussian elevation distribution for each target cell. The paper's two load-bearing additions are semantic conditioning, which feeds fused LiDAR and camera features at context and target locations so the model can extrapolate into unobserved regions, and ball-query attention, which restricts self- and cross-attention to keys inside an $\epsilon$-ball around each query in the bird's-eye-view grid. A non-parametric Bayesian update aggregates image BEV features over time without additional training. Training maximizes an evidence lower bound, and the attention is scaled dot-product over local neighborhoods; this locality is what makes the $100\,\text{m}\times 100\,\text{m}$ grid tractable for onboard computing.

What would settle it

Re-run the three sequences with an explicit temporally disjoint held-out split, multiple random seeds, and a surveyed ground truth such as RTK-GPS or dense post-processed scans instead of the 300-scan minimum height; if the elevation margin over the LiDAR baseline collapses or the min-height labels deviate from the surveyed surface near ditches and vegetation, the central claim is falsified.

Watch

Extended reading notes

Core claim

The central claim is that an Attentive Neural Process, extended with semantic conditioning and local ball-query attention, precisely estimates sharp elevation changes and quantifies predictive uncertainty without losing elevation accuracy. Observed LiDAR ground heights serve as context points, and context and target points are augmented with fused LiDAR/camera semantic features, so the model can interpolate and extrapolate into unobserved BEV cells. Ball-query attention replaces global attention with attention over observations inside a small neighborhood, cutting GFLOPs by 17% and peak GPU memory by 78% while keeping local sharpness. On the CA Hills, Mojave Desert, and Ellensburg sequences, the paper reports the lowest total elevation and slope errors among all baselines, improves curvature over the neural baselines, and keeps the accuracy drop small when transferring from one site to another; its predictive uncertainty also outperforms an uncertainty-trained neural baseline in the reported transfer scenarios.

Load-bearing premise

The comparison rests on two unstated evaluation assumptions: that the minimum LiDAR height over a 300-scan window is an unbiased ground-truth elevation label for every BEV cell, and that training and evaluation frames are separated in time; because the paper does not describe its train/test splits, seeds, or repeated runs, the reported margins over the next-best baseline could be run-to-run noise.

Editorial extensions

If this is right

  • Off-road planners can query both a height and a variance for every BEV cell, so high-risk negative obstacles and low-confidence regions can be flagged rather than smoothed away.
  • Ball-query attention makes the neural elevation model deployable at scale, reducing floating-point operations by 17%, peak GPU memory by 78%, and inference time by 36% relative to global attention.
  • Uncertainty estimation no longer has to trade away elevation accuracy, because the same network produces point predictions and a predictive distribution.
  • Cross-terrain generalization improves because semantic conditioning keeps the accuracy drop small when a model trained on one site is applied to another terrain type.

Reading between the lines

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

  • A clean test of the paper's attribution of the gains would be to give the strongest LiDAR-only baseline the same fused semantic features; if that baseline then matches the reported accuracy, semantic conditioning rather than the Neural Process objective is the load-bearing component.
  • The same semantic-conditioned Neural Process construction transfers naturally to other continuous spatial fields in robotics, such as traversability cost or ground-friction estimates, where sparse observations must be extrapolated to unobserved cells.
  • The paper's own ablation shows that temporally aggregating image features adds only a small elevation gain, which suggests most of the temporal improvement comes from aggregating LiDAR points; making the camera stream load-bearing may require the learned confidence weighting the paper lists as future work.
  • If the reported margins survive a temporally disjoint evaluation, semantic-conditioned Neural Processes could become the default backend for BEV elevation estimation, with the ball radius as an explicit accuracy-versus-compute dial.
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 / 5 minor

Summary. The manuscript proposes a semantic-conditioned Attentive Neural Process (ANP) for bird's-eye-view terrain elevation estimation from LiDAR and camera inputs. The method combines LiDAR and image BEV features, replaces global attention with a local ball-query attention mechanism, and temporally aggregates image features with a non-parametric Bayesian update. The authors evaluate on three off-road sequences (CA Hills, Mojave Desert, Ellensburg) and report lower elevation, slope, and curvature errors than GP and neural baselines, along with uncertainty metrics and ablations.

Significance. If the reported comparisons hold, the work would be a useful step for off-road terrain modeling: it demonstrates a practical way to obtain uncertainty estimates from Neural Processes without losing sharp elevation detail, and the ball-query attention is a sensible efficiency mechanism for onboard computation. Strengths include the explicit derivation of the Bayesian temporal update in Appendix A, the inclusion of modality and attention-radius ablations, and the computational cost analysis showing clear memory savings. The main weakness is that the empirical evidence as reported does not yet establish the headline outperformance claim, because the evaluation protocol is underspecified and several reported margins are very small.

major comments (4)
  1. [§3.1, Tables 1 and 2] The paper never states how the three sequences are split into training, validation, and test sets. This is load-bearing because the ground-truth elevation for frame t is the minimum height over scans (t−150, t+149); if training and evaluation frames are temporally interleaved, test labels can depend on observations from training frames, and cells labeled "unobserved" at time t may have been observed during training. Table 7 explicitly distinguishes "random samples" from "first n% frames," but it does not say which protocol produced the results in Tables 1 and 2. Please specify the split (e.g., contiguous time intervals), the number of frames in each split, and confirm that no test frame's 300-scan label window overlaps any training frame.
  2. [Tables 1 and 2] All results appear to come from a single training run, with no seeds, error bars, or significance tests. Several headline margins are very small: on Mojave, Ours-TA total elevation MAE is 0.225 m versus Fusion-TA 0.229 m (4 mm); on CA Hills, Ours-TA is 0.512 m versus 0.527 m. For models of this capacity, these differences are within plausible run-to-run or frame-to-frame noise. Please report means and standard deviations over at least three seeds, or paired per-frame differences, and state whether the improvements are consistent across seeds and time intervals.
  3. [§3.3, Table 1, abstract] The abstract and Section 1 claim the method estimates elevation "without losing elevation accuracy" and "without compromising prediction accuracy." However, Table 1 shows that temporal aggregation consistently worsens observed-region elevation MAE relative to the single-frame model: 0.304 vs 0.280 on CA Hills, 0.154 vs 0.131 on Mojave, and 0.299 vs 0.282 on Ellensburg. The paper acknowledges this observed-region increase in §3.3, but the headline claim needs to be disambiguated: is the accuracy claim about total MAE only, and is the observed-region degradation considered an acceptable trade-off? Please state which metric the claim refers to and discuss this trade-off explicitly.
  4. [§2.2 and Appendix A] The Bayesian temporal update in Eq. (3) relies on treating the normalized density of lifted image pixels as the probability that a BEV feature is correct, together with a conditional-independence and uniform-prior assumption. These assumptions are stated explicitly in Appendix A, which is good, but their validity is not empirically checked. A calibration analysis (e.g., density vs. observed correctness over held-out frames) would strengthen the claim that this is a principled Bayesian update rather than a heuristic weighting rule.
minor comments (5)
  1. [§3.2] Please define the slope and curvature MAE metrics more precisely; for example, state whether slope error is measured in percentage points or as a relative error, and describe the Laplacian filter used for curvature.
  2. [§2.2 and Appendix A] Equation (3) and Eq. (6) are identical; please unify the notation to avoid duplication.
  3. [Abstract and Table 5] The abstract says the method reduces "computational complexity" by 17%, but Table 5 reports a 17% reduction in GFLOPs; asymptotic complexity is not analyzed. I suggest rewording to "reduces GFLOPs by 17%" or reporting the actual complexity bounds.
  4. [§C.2, Table 7] The sample-efficiency ablation is informative but would be more useful if it stated which split protocol (random vs. first n% frames) corresponds to the main results in Table 1.
  5. [Section 5] The limitation section appropriately notes the lack of validation beyond 51.2 m and the need for uncertainty calibration; please also mention there that the observed-region elevation accuracy degrades under temporal aggregation, since that is a known limitation of the current version.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the derivation chain is self-contained and external benchmarks anchor all claims.

full rationale

The paper's central claims are empirical comparisons against external ground-truth elevations, not against the model's own outputs. The elevation label is defined as the minimum height over a 300-scan window of LiDAR observations (Section 3.1), and evaluation metrics (MAE of elevation, slope, and curvature) are computed with respect to this external ground truth. No fitted constant is renamed as a prediction: the temporal Bayesian update in Eq. (3) follows from stated Bayes' rule and conditional-independence assumptions in Appendix A, and the ball-query attention FLOPs reduction is measured, not defined by construction. The adoption of Attentive Neural Processes as a base framework cites external work (Kim et al., 2019), and the semantic encoders (DinoV2, Asymmetric 3D U-Net) are independent published components. The paper does cite prior work by overlapping authors (TerrainNet, BEVNet, V-STRONG), but these citations are used for baselines and related context, not as the load-bearing justification for the paper's own contributions. The absence of train/test split details, seeds, or repeated runs (Tables 1 and 2) is a legitimate evaluation-robustness concern, but it is not a circularity: the reported margins are externally anchored and could be re-measured on a properly held-out split. No equation is equivalent to its inputs by construction, and no claim reduces to a self-citation chain.

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

No new physical entity is postulated; the model uses a latent variable inherited from Attentive Neural Processes. The main uncharged premises are the label-generation assumptions and the Bayesian-update heuristic for image features. The empirical claim also depends on a fair train/test split, which is not described.

free parameters (5)
  • Neural network weights = Trained on the three datasets
    All reported performance is produced by a fitted model; the paper does not fully specify optimization schedules, loss weights, or split construction.
  • Ball radius epsilon = 2.0 m
    Selected after a CA Hills ablation (Table 4); affects attention locality and all reported in-domain numbers.
  • BEV resolution R = 0.4 m
    Discretization chosen by hand; defines context and target points and all evaluation metrics.
  • Ego-centric range = 51.2 m
    Fixed crop limit; the extrapolation claims are bounded to this range, as the paper acknowledges in its limitation section.
  • Maximum context points during training = 7,000
    Chosen for memory reasons; affects the training distribution and may influence extrapolation behavior.
assumptions (5)
  • domain assumption LiDAR minimum-height in each BEV cell approximates ground elevation.
    Invoked in the problem formulation (Section 2.1) to define context elevations from raw LiDAR hits.
  • domain assumption Ground truth generated by taking the minimum over 300 aggregated LiDAR scans is the true terrain elevation.
    Section 3.1 defines the evaluation label this way; if this minimum is biased under vegetation or steep slopes, all reported errors are affected.
  • ad hoc to paper The normalized density of lifted image pixels is a valid proxy for the probability that the image BEV feature is correct.
    Used in Section 2.2 and Appendix A to justify the temporal Bayesian update of image features.
  • ad hoc to paper Conditional independence of current and previous image features given correctness, plus a uniform prior, holds for the Bayesian temporal update.
    Appendix A derives Equation (3) using these assumptions; they are simplifying choices not validated against data.
  • domain assumption Semantic features from DinoV2 and the asymmetric 3D U-Net, after projection to BEV, carry useful information for elevation extrapolation.
    Section 2.2 relies on this transfer for interpolation and extrapolation gains in unobserved regions.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Uncertainty-aware Accurate Elevation Modeling for Off-road Navigation via Neural Processes." pith.science (2026). https://pith.science/paper/E4YLA7WP

@misc{pith2026250803890,
  author       = {Pith},
  title        = {Pith review of: Uncertainty-aware Accurate Elevation Modeling for Off-road Navigation via Neural Processes},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/E4YLA7WP}},
  note         = {Machine review of arXiv:2508.03890}
}
read the original abstract

Terrain elevation modeling for off-road navigation aims to accurately estimate changes in terrain geometry in real-time and quantify the corresponding uncertainties. Having precise estimations and uncertainties plays a crucial role in planning and control algorithms to explore safe and reliable maneuver strategies. However, existing approaches, such as Gaussian Processes (GPs) and neural network-based methods, often fail to meet these needs. They are either unable to perform in real-time due to high computational demands, underestimating sharp geometry changes, or harming elevation accuracy when learned with uncertainties. Recently, Neural Processes (NPs) have emerged as a promising approach that integrates the Bayesian uncertainty estimation of GPs with the efficiency and flexibility of neural networks. Inspired by NPs, we propose an effective NP-based method that precisely estimates sharp elevation changes and quantifies the corresponding predictive uncertainty without losing elevation accuracy. Our method leverages semantic features from LiDAR and camera sensors to improve interpolation and extrapolation accuracy in unobserved regions. Also, we introduce a local ball-query attention mechanism to effectively reduce the computational complexity of global attention by 17\% while preserving crucial local and spatial information. We evaluate our method on off-road datasets having interesting geometric features, collected from trails, deserts, and hills. Our results demonstrate superior performance over baselines and showcase the potential of neural processes for effective and expressive terrain modeling in complex off-road environments.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

53 extracted references · 43 canonical work pages

  1. [1]

    Guizilini and F

    V . Guizilini and F. Ramos. Variational hilbert regression for terrain modeling and trajectory optimization. The International Journal of Robotics Research, 38(12-13):1375–1387, 2019

    work page 2019

  2. [2]

    Chung, G

    C. Chung, G. Georgakis, P. Spieler, C. Padgett, A. Agha, and S. Khattak. Pixel to elevation: Learning to predict elevation maps at long range using images for autonomous offroad naviga- tion. IEEE Robotics and Automation Letters, 2024

    work page 2024

  3. [3]

    M. Oh, G. Shin, S. Jang, S. Lee, D. Lee, W. Song, B. Yu, H. Lim, J. Lee, and H. Myung. B- tms: Bayesian traversable terrain modeling and segmentation across 3d lidar scans and maps for enhanced off-road navigation. In 2024 IEEE Intelligent Vehicles Symposium (IV) , pages 200–205. IEEE, 2024

    work page 2024

  4. [4]

    Datar, C

    A. Datar, C. Pan, M. Nazeri, A. Pokhrel, and X. Xiao. Terrain-attentive learning for efficient 6- dof kinodynamic modeling on vertically challenging terrain. In 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 5438–5443. IEEE, 2024

    work page 2024

  5. [5]

    A. Dai, S. Gupta, and G. Gao. Neural elevation models for terrain mapping and path planning. arXiv preprint arXiv:2405.15227, 2024

    arXiv 2024

  6. [6]

    A. L. Montealegre, M. T. Lamelas, and J. De La Riva. Interpolation routines assessment in als- derived digital elevation models for forestry applications. Remote sensing, 7(7):8631–8654, 2015

    work page 2015

  7. [7]

    H. Feng, X. Xu, and L. De Floriani. Implicitterrain: a continuous surface model for terrain data analysis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 899–909, 2024

    work page 2024

  8. [8]

    C. A. Dimmig, K. C. Wolfe, B. Woosley, M. Kobilarov, and J. Moore. Uncertainty-aware planning for heterogeneous robot teams using dynamic topological graphs and mixed-integer programming. arXiv preprint arXiv:2310.08396, 2023

    arXiv

Show all 53 references
  1. [9]

    Fahnestock, E

    E. Fahnestock, E. Fuentes, P. R. Osteen, S. Ancha, and N. Roy. Learning semantic traversabil- ity priors using diffusion models for uncertainty-aware global path planning. In ICRA 2024 Workshop on Resilient Off-road Autonomy, 2024

    2024

  2. [10]

    W. Shao, J. Xu, Z. Cao, H. Wang, and J. Li. Uncertainty-aware prediction and application in planning for autonomous driving: Definitions, methods, and comparison. arXiv preprint arXiv:2403.02297, 2024

    2024 arXiv

  3. [11]

    Schmittle, R

    M. Schmittle, R. Baijal, N. Hatch, R. Scalise, M. G. Castro, S. Talia, K. Khetarpal, B. Boots, and S. Srinivasa. Long range navigator (lrn): Extending robot planning horizons beyond metric maps. arXiv preprint arXiv:2504.13149, 2025

    2025 arXiv

  4. [12]

    M. Seeger. Gaussian processes for machine learning. International journal of neural systems, 14(02):69–106, 2004

    2004

  5. [13]

    Vasudevan, F

    S. Vasudevan, F. Ramos, E. Nettleton, and H. Durrant-Whyte. Gaussian process modeling of large-scale terrain. Journal of Field Robotics, 26(10):812–840, 2009

    2009

  6. [14]

    Torroba, C

    I. Torroba, C. I. Sprague, and J. Folkesson. Fully-probabilistic terrain modelling with stochastic variational gaussian process maps. arXiv preprint arXiv:2203.10893, 2022

    2022 arXiv

  7. [15]

    Viset, R

    F. Viset, R. Helmons, and M. Kok. Spatially scalable recursive estimation of gaussian process terrain maps using local basis functions. IEEE Transactions on Signal Processing, 2025

    2025

  8. [16]

    A. H. Chang, C. M. Hubicki, J. J. Aguilar, D. I. Goldman, A. D. Ames, and P. A. Vela. Learn- ing terrain dynamics: A gaussian process modeling and optimal control adaptation framework applied to robotic jumping. IEEE Transactions on Control Systems Technology, 29(4):1581– 1596...

    2021

  9. [17]

    T. Lang, C. Plagemann, and W. Burgard. Adaptive non-stationary kernel regression for terrain modeling. In Robotics: Science and Systems, volume 6. Citeseer, 2007

    2007

  10. [18]

    Vasudevan, F

    S. Vasudevan, F. Ramos, E. Nettleton, and H. Durrant-Whyte. Heteroscedastic gaussian pro- cesses for data fusion in large scale terrain modeling. In 2010 IEEE International Conference on Robotics and Automation, pages 3452–3459. IEEE, 2010

    2010

  11. [19]

    Vasudevan, F

    S. Vasudevan, F. Ramos, E. Nettleton, and H. Durrant-Whyte. Non-stationary dependent gaus- sian processes for data fusion in large-scale terrain modeling. In 2011 IEEE International Conference on Robotics and Automation, pages 1875–1882. IEEE, 2011

    2011

  12. [20]

    Zhang, G

    Z. Zhang, G. Dai, D. Wang, and M. I. Jordan. Bayesian generalized kernel models. In Y . W. Teh and M. Titterington, editors, Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, volume 9 of Proceedings of Machine Learning Research, ...

    2010

  13. [21]

    T. Shan, J. Wang, B. Englot, and K. Doherty. Bayesian generalized kernel inference for terrain traversability mapping. In Conference on Robot Learning, pages 829–838. PMLR, 2018

    2018

  14. [22]

    Doherty, T

    K. Doherty, T. Shan, J. Wang, and B. Englot. Learning-aided 3-d occupancy mapping with bayesian generalized kernel inference. IEEE Transactions on Robotics, 35(4):953–966, 2019. doi:10.1109/TRO.2019.2912487

    2019

  15. [23]

    H. Xue, H. Fu, L. Xiao, Y . Fan, D. Zhao, and B. Dai. Traversability analysis for autonomous driving in complex environment: A lidar-based terrain modeling approach. Journal of Field Robotics, 40(7):1779–1803, 2023

    2023

  16. [24]

    Shaban, X

    A. Shaban, X. Meng, J. Lee, B. Boots, and D. Fox. Semantic terrain classification for off-road autonomous driving. In Conference on Robot Learning, pages 619–629. PMLR, 2022

    2022

  17. [25]

    X. Meng, N. Hatch, A. Lambert, A. Li, N. Wagener, M. Schmittle, J. Lee, W. Yuan, Z. Q. Chen, S. Deng, G. Okopal, D. Fox, B. Boots, and A. Shaban. Terrainnet: Visual modeling of complex terrain for high-speed, off-road navigation. In K. E. Bekris, K. Hauser, S. L. Herbert, and ...

    2023

  18. [26]

    S. Jung, J. Lee, X. Meng, B. Boots, and A. Lambert. V-strong: Visual self-supervised traversability learning for off-road navigation. In 2024 IEEE International Conference on Robotics and Automation (ICRA), pages 1766–1773. IEEE, 2024

    2024

  19. [27]

    Garnelo, J

    M. Garnelo, J. Schwarz, D. Rosenbaum, F. Viola, D. J. Rezende, S. Eslami, and Y . W. Teh. Neural processes. arXiv preprint arXiv:1807.01622, 2018

    2018 arXiv

  20. [28]

    H. Kim, A. Mnih, J. Schwarz, M. Garnelo, A. Eslami, D. Rosenbaum, O. Vinyals, and Y . W. Teh. Attentive neural processes. arXiv preprint arXiv:1901.05761, 2019

    1901 arXiv

  21. [29]

    J. Lee, Y . Lee, J. Kim, E. Yang, S. J. Hwang, and Y . W. Teh. Bootstrapping neural processes. Advances in neural information processing systems, 33:6606–6615, 2020

    2020

  22. [30]

    Gordon, W

    J. Gordon, W. P. Bruinsma, A. Y . Foong, J. Requeima, Y . Dubois, and R. E. Turner. Convolu- tional conditional neural processes. arXiv preprint arXiv:1910.13556, 2019

    1910 arXiv

  23. [31]

    Vaswani, N

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. u. Kaiser, and I. Polosukhin. Attention is all you need. In I. Guyon, U. V . Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett, editors, Ad- vances in Neural Information Process...

  24. [32]

    X. Zhu, H. Zhou, T. Wang, F. Hong, Y . Ma, W. Li, H. Li, and D. Lin. Cylindrical and asym- metrical 3d convolution networks for lidar segmentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 9939–9948, 2021

    2021

  25. [33]

    Oquab, T

    M. Oquab, T. Darcet, T. Moutakanni, H. V o, M. Szafraniec, V . Khalidov, P. Fernandez, D. Haz- iza, F. Massa, A. El-Nouby, et al. Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193, 2023

    2023 arXiv

  26. [34]

    D. P. Kingma, M. Welling, et al. Auto-encoding variational bayes, 2013

    2013

  27. [35]

    Philion and S

    J. Philion and S. Fidler. Lift, splat, shoot: Encoding images from arbitrary camera rigs by implicitly unprojecting to 3d. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XIV 16, pages 194–210. Springer, 2020

    2020

  28. [36]

    Robbiano, E

    C. Robbiano, E. K. Chong, M. R. Azimi-Sadjadi, L. L. Scharf, and A. Pezeshki. Bayesian learning of occupancy grids. IEEE Transactions on Intelligent Transportation Systems, 23(2): 1073–1084, 2020

    2020

  29. [37]

    Nguyen and A

    T. Nguyen and A. Grover. Transformer neural processes: Uncertainty-aware meta learning via sequence modeling. In International Conference on Machine Learning, ICML 2022, 17-23 July 2022, Baltimore, Maryland, USA, volume 162 of Proceedings of Machine Learning Research, pages 16...

    2022

  30. [38]

    URL https://www.polaris.com/en-us/off-road/rzr/

    Polaris RZR. URL https://www.polaris.com/en-us/off-road/rzr/

  31. [39]

    W. Hess, D. Kohler, H. Rapp, and D. Andor. Real-time loop closure in 2d lidar slam. In 2016 IEEE international conference on robotics and automation (ICRA) , pages 1271–1278. IEEE, 2016

    2016

  32. [40]

    Damianou and N

    A. Damianou and N. D. Lawrence. Deep gaussian processes. In Artificial intelligence and statistics, pages 207–215. PMLR, 2013

    2013

  33. [41]

    Ramos and L

    F. Ramos and L. Ott. Hilbert maps: Scalable continuous occupancy mapping with stochastic gradient descent. The International Journal of Robotics Research, 35(14):1717–1730, 2016

    2016

  34. [42]

    Snelson and Z

    E. Snelson and Z. Ghahramani. Sparse gaussian processes using pseudo-inputs. Advances in neural information processing systems, 18, 2005

    2005

  35. [43]

    Sheth, Y

    R. Sheth, Y . Wang, and R. Khardon. Sparse variational inference for generalized gp models. In International Conference on Machine Learning, pages 1302–1311. PMLR, 2015

    2015

  36. [44]

    M. Titsias. Variational learning of inducing variables in sparse gaussian processes. InArtificial intelligence and statistics, pages 567–574. PMLR, 2009

    2009

  37. [45]

    D. Levi, L. Gispan, N. Giladi, and E. Fetaya. Evaluating and calibrating uncertainty prediction in regression tasks. Sensors, 22(15):5540, 2022

    2022

  38. [46]

    S. Jha, D. Gong, X. Wang, R. E. Turner, and L. Yao. The neural process family: Survey, applications and perspectives. arXiv preprint arXiv:2209.00517, 2022

    2022 arXiv

  39. [47]

    Senanayake and F

    R. Senanayake and F. Ramos. Bayesian hilbert maps for dynamic continuous occupancy map- ping. In Conference on Robot Learning, pages 458–471. PMLR, 2017

    2017

  40. [48]

    C. L. Gentil, C. Pradalier, and T. D. Barfoot. Towards efficient occupancy mapping via gaussian process latent field shaping. arXiv preprint arXiv:2506.13640, 2025

    2025 arXiv

  41. [49]

    Achey, A

    L. Achey, A. Reed, B. Crowe, B. Hayes, and C. Heckman. Robust robotic exploration and mapping using generative occupancy map synthesis. arXiv preprint arXiv:2506.20049, 2025. 20

    2025

  42. [50]

    J. R. Gardner, G. Pleiss, D. Bindel, K. Q. Weinberger, and A. G. Wilson. Gpytorch: Black- box matrix-matrix gaussian process inference with gpu acceleration. In Advances in Neural Information Processing Systems, 2018

    2018

  43. [51]

    Tan and Q

    M. Tan and Q. Le. EfficientNet: Rethinking model scaling for convolutional neural networks. In K. Chaudhuri and R. Salakhutdinov, editors, Proceedings of the 36th International Con- ference on Machine Learning , volume 97 of Proceedings of Machine Learning Research , pages 610...

    2019

  44. [52]

    D. P. Kingma and J. Ba. Adam: A method for stochastic optimization. In Y . Bengio and Y . LeCun, editors, 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings , 2015. URL http:// arxiv.org/abs/1412...

    2015 arXiv

  45. [2017]

    URL https://proceedings.neurips.cc/paper_files/paper/2017/file/ 3f5ee243547dee91fbd053c1c4a845aa-Paper.pdf. 19

    2017

Pith tools

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