Pith. sign in

REVIEW 5 major objections 7 minor 42 references

Radiant: Large-scale 3D Gaussian Rendering based on Hierarchical Framework

T0 review · 5 major / 7 minor · reviewed 2026-08-11 · deepseek-v4-flash

Pith's one-line read Radiant: a hierarchical cloud-edge-device framework for large-scale 3D Gaussian scene reconstruction that claims up to 25.7% better reconstruction quality and up to 79.6% lower end-to-end latency than current distributed baselines.

desk verdict A genuine but overclaimed systems paper: the hierarchical 3DGS framework is new and mostly works, but the headline quality/latency gains come from weak baselines and unreported profiling fits. read the letter →

arxiv 2412.05546 v1 pith:A5XAELIF submitted 2024-12-07 cs.CV cs.DC

classification cs.CVcs.DC
keywords 3DGaussianSplattingdistributedreconstructioncloud-edge-devicearchitectureworkloadpartitioningregionplanningmodelaggregationsyntheticviewslatencyminimization
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper proposes Radiant, a three-layer cloud-edge-device architecture for training 3D Gaussian Splatting models on very large scenes without gathering all images in one place. Its central claim is that intelligent geographic partitioning of the scene among edge servers, together with workload allocation among heterogeneous devices, can nearly eliminate stragglers and cut end-to-end latency by up to 79.6% while improving reconstruction quality by up to 25.7% over cloud-device federated baselines. The authors argue that existing distributed 3DGS methods ignore system heterogeneity and that their hierarchical approach achieves quality comparable to centralized training while preserving privacy and scalability. If true, this makes large-scale 3D reconstruction practical on ordinary edge hardware.

What carries the argument

The argument runs on three mechanisms. Adaptive Region Planning (ARP) initializes non-overlapping regions as a Voronoi diagram of edge locations, then iteratively expands or shrinks each region's boundary by a distance step toward a threshold so that every edge's estimated completion time is close to the average. Resource-aware Task Partitioning (RTP) solves the min-max assignment of camera positions to devices, using offline-fitted functions $F_n$ (training time vs image count), $G$ (initialization time), and $H$ (model size) and each device's bandwidth to predict completion times. The model aggregation step takes each device's trained Gaussians, renders synthetic views from the device's own camera poses, concatenates all Gaussians into one edge model, and retrains for a small number of epochs; the same procedure fuses edge models at the cloud. The synthetic-view retraining repairs boundary discontinuities without ever uploading raw images.

What would settle it

Run the same scene and system configuration but replace the offline-fitted time functions with the actual measured per-device training and communication times; if the predicted completion times deviate from the measured ones by a large margin, or if the ARP boundary iteration fails to converge to equal completion times, the central latency claim is falsified.

Watch

Extended reading notes

Core claim

Radiant's discovery is that the bottleneck in distributed 3D Gaussian Splatting is not raw compute but load imbalance: because devices differ in GPU power, bandwidth, and camera counts, a naive even split makes the slowest region set the wall-clock time. The paper shows that by (1) partitioning the scene with a Voronoi-based adaptive region planning algorithm that repeatedly moves boundaries to equalize predicted completion times, (2) allocating camera positions to devices by solving a min-max load balancing problem with offline-profiled training and communication time functions, and (3) fusing device models at the edge by concatenating them and retraining a few epochs on synthetic views, one can obtain a global model that is both faster to build and better at region boundaries. The reported numbers: Radiant improves PSNR by up to 25.7% over the federated baseline and reduces end-to-end latency by up to 79.6% on two large outdoor scenes.

Load-bearing premise

The load-bearing premise is that the offline-fitted functions for training time, initialization time, and model size accurately predict completion times on the deployed devices; if those profiles are wrong, the region boundaries and camera allocations that ARP and RTP compute will not actually balance the load, and the claimed latency reduction would not follow.

Editorial extensions

If this is right

  • Large outdoor scenes spanning about 100,000 square meters can be reconstructed on four heterogeneous edge regions in roughly 1,500 to 2,100 seconds on the testbed, versus thousands of seconds for a federated baseline.
  • Raw images and precise device locations never leave the device, so the framework preserves privacy while still yielding a globally consistent model.
  • New scenes can be added as new regions without retraining existing ones, so the system scales by adding edges and devices.
  • The synthetic-view aggregation step makes quality degrade only mildly as the number of regions grows, unlike direct boundary cutting and merging.

Reading between the lines

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

  • Editorial inference: the same ARP/RTP load-balancing recipe could be applied to other distributed neural rendering systems, since the profiled functions only need training time, initialization time, and model size as a function of camera count.
  • Editorial inference: because the aggregation step uses synthetic views rendered from the device models, it could in principle be iterated, and a second round of synthetic-view retraining at the cloud might further smooth inter-edge boundaries; the paper does not explore this.
  • Editorial inference: the reported latency gains may depend on the accuracy of the offline profiles, so an online estimator that updates the fitted functions from observed runtimes would make the scheme robust to changing network conditions.
  • Editorial inference: an ablation that compares Radiant's partitioning against a random partition using the same aggregation step would isolate how much of the quality gain comes from load balancing versus from the synthetic-view retraining.
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

5 major / 7 minor

Summary. The paper presents Radiant, a hierarchical cloud-edge-device framework for distributed 3D Gaussian Splatting reconstruction of large scenes. It introduces Adaptive Region Planning (ARP) to partition the scene among edges and Resource-aware Task Partitioning (RTP) to allocate cameras to devices, both based on offline-profiled completion-time functions. Device models are aggregated at edges by concatenating models and retraining on synthetic views rendered from the device models, and the resulting edge models are fused at the cloud. Experiments on the Mill 19 Rubble and Building scenes with a heterogeneous testbed report PSNR/SSIM/LPIPS and end-to-end latency, claiming up to 25.7% quality improvement and up to 79.6% latency reduction over Fed3DGS, along with internal ablations on aggregation and partitioning.

Significance. The idea of treating region and task partitioning as a latency-minimization problem in a three-tier cloud-edge-device system is a worthwhile contribution, and the testbed experiments are a useful step toward practical distributed 3DGS. The synthetic-view retraining aggregation is simple and appears effective in the internal ablation (Table III). However, the headline claims are not yet fully supported: the quality comparison omits the strongest recent baselines (CityGaussian, DOGaussian) and is worse than VastGaussian, and the latency gain over Fed3DGS is mostly due to Fed3DGS's serial fusion rather than the proposed partitioning. The offline profile functions that drive ARP/RTP are not reported with any fidelity or sensitivity analysis, and the per-device SfM initialization leaves the coordinate-alignment problem unaddressed. With these issues fixed, the system could be a solid contribution.

major comments (5)
  1. [Table II / Sec. IV-C] The claim that 'Radiant improved reconstruction quality by up to 25.7%' is measured against Drone-NeRF on Rubble (24.53 vs 19.51 PSNR), the weakest baseline in Table II. Against VastGaussian, Radiant-system2 is worse on Rubble (24.53 vs 25.20) and slightly worse on Building (21.66 vs 21.80). Since CityGaussian [8] and DOGaussian [9] are cited in Section V as distributed 3DGS methods but are not included in Table II, the statement that Radiant outperforms state-of-the-art methods is not supported. The authors should either add these baselines or revise the claim to 'comparable quality with privacy and scalability benefits.'
  2. [Sec. IV-D, Eqs. (13)-(16)] The 79.6% latency reduction over Fed3DGS is attributable to Fed3DGS fusing all device models sequentially on a single machine, as the paper itself states in Section IV-D; the algorithm-specific gain of ARP+RTP over even partitioning is 19.27% (Rubble) and 20.99% (Building). This gain rests entirely on the offline-fitted functions F_n, G, H in Eqs. (13)-(16), but the paper gives no fitted forms, fitting error, or sensitivity analysis. Because model size H(|C_n|) depends on scene content and densification, not just image count, the profiles may not transfer across scenes; without this evidence the solutions to Eqs. (9) and (17) are not shown to be latency-minimizing. Please report the fitted functions, their fit quality, and an ablation in which the partition is computed with perturbed profiles.
  3. [Sec. III-A, III-C, III-D] The workflow says each device independently collects images and initializes the 3D Gaussian points by SfM (Section III-A, step b). No mechanism is described for aligning the per-device SfM reconstructions into a common coordinate frame before models are concatenated or before synthetic views are rendered for retraining in Section III-D. Without a common frame, the operations 'concatenate all device models' (Section III-D) and 'cut out the 3D Gaussians outside the boundary lines' are undefined. The paper must specify how global camera poses or a global registration is obtained (e.g., GPS/IMU priors, shared SfM, or post-hoc alignment) and how this is consistent with the stated privacy guarantees.
  4. [Algorithm 1 / Sec. III-B] The ARP heuristic moves boundaries by a fixed distance d until the residual threshold T_thres is met, but the paper does not state the values of d and T_thres, nor does it provide any convergence or sensitivity analysis. The statement that ARP 'can quickly converge through rapid iterations' is supported only by the four curves in Fig. 13. Please report d and T_thres, and show how the resulting end-to-end latency depends on them.
  5. [Sec. III-C, Eq. (17)] After stating that G, F, and H are known, the paper says the number of camera positions per device 'can be directly calculated' but gives no explicit solution or algorithm for problem (17). Without this, the RTP allocation is underspecified and the ARP completion-time estimates are not reproducible. Please provide the closed-form or algorithmic solution used to allocate |C_n|.
minor comments (7)
  1. [Eq. (7)] Equation (7) uses T^n_aggre but should be T^m_aggre, since it is the aggregation time at edge m.
  2. [Sec. II-C] The text says 'The overall latency is illustrated in Fig. 2b' but the referenced figure is Fig. 3b.
  3. [Sec. IV-B4 and IV-D] Section IV-B4 misspells 'VastGaussian' as 'Vast-Faussian' and Section IV-D misspells 'Fed3DGS' as 'Fededgs'; please correct these.
  4. [Table II] The legend for the Scalability and Privacy symbols ('!' and '%') is missing; please define the symbols in the table caption or in the text.
  5. [Sec. IV] The paper reports single-run latency and quality numbers; please state the number of runs or provide error bars to indicate variance.
  6. [Sec. IV] No reproducibility statement is provided; please release code, fitted profile functions, or at least the exact hyperparameters d and T_thres to allow the experiments to be repeated.
  7. [Sec. I, III-A] The scalability claim that new scenes can be added as new edges without impacting already trained regions is not validated experimentally; consider an experiment with varying numbers of edges and devices.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the headline quality and latency numbers are measured on a testbed, and the offline-fitted profile functions are explicit optimization inputs rather than predictions derived from the target result.

full rationale

Walking the claimed derivation chain: (1) The workload partitioning problems (Eqs. 9 and 17) are solved using offline-profiled functions Fn, G, and H (Eqs. 13-16), which the paper explicitly states can be 'profiled offline' from device measurements. These functions are inputs to the scheduler, not quantities that are then presented as newly predicted results. (2) The paper's headline numbers (Table II for PSNR/SSIM/LPIPS and Fig. 9 for end-to-end latency) are obtained by actually running the Radiant prototype on a heterogeneous testbed, not by evaluating the fitted equations. Thus the fitted profiles are not dressed up as predictions of the final metrics. (3) The synthetic-view aggregation step renders Ir = R(Cn, Gn) and retrains the concatenated model on those rendered views; this is a self-distillation training procedure whose boundary-quality benefit is empirically measured in Table III and Fig. 12, not entailed by definition. (4) The self-citations [3] and [6] include overlapping authors (H. Li, Y. Dai), but they appear only in the related-work enumeration of 3DGS variants and are not load-bearing; there is no uniqueness theorem or first-principles premise imported from those citations. The lack of reported fitted forms, fitting error, and sensitivity analysis for Fn, G, and H is a reproducibility and robustness concern, not circularity. No load-bearing step in the paper's derivation reduces to its own input by construction.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

No new physical entities. The paper's contribution is a system design; its free parameters are the profiling functions and hyperparameters that drive the optimization. The main unstated support comes from fitted profiles and the synthetic-view retraining assumption.

free parameters (6)
  • Fn (training time profile for device n) = not reported
    Fitted offline from Fig. 2a; used in Eq. (16) to predict per-device training time from camera count.
  • G (initialization time profile) = not reported
    Fitted offline; Eq. (13) maps camera count to SfM initialization time.
  • H (model size profile) = not reported
    Fitted offline from Fig. 2b; Eq. (14) maps camera count to model size, then used for communication time in Eq. (15).
  • d (boundary movement step) = not reported
    ARP hyperparameter in Algorithm 1, line 9, controlling how far boundaries move per iteration; chosen by hand, not justified.
  • T_thres (termination threshold) = not reported
    ARP stopping criterion in Algorithm 1; chosen by hand, not justified.
  • Retraining epochs for aggregation = 10
    Model aggregation retrains edge and cloud models for 10 epochs (Section IV-B3); this is a free choice affecting fusion quality.
assumptions (5)
  • standard math 3D Gaussian splatting forward model (Eqs. 1-4) is a valid scene representation
    Background from Kerbl et al. [1]; accepted rendering model.
  • domain assumption Completion time for an edge is the sum of training, aggregation, and communication times, and devices within an edge train in parallel (Eqs. 5-8)
    Assumes no overlap or contention between computation and communication; ignores straggler variance beyond the max-device time.
  • ad hoc to paper Offline-fitted profiling functions Fn, G, H generalize to the test scenes and remain valid during deployment
    The optimization in Eqs. (9) and (17) relies on these fitted functions; no validation of their accuracy on Mill 19 or under actual network conditions.
  • ad hoc to paper Retraining on synthetic views rendered from device models improves boundary quality without real images
    Core of aggregation in Section III-D; ablation Table III supports it on two scenes, but assumes synthetic views carry enough information to correct boundary artifacts.
  • domain assumption Non-overlapping region partition (Eq. 11) preserves privacy and coverage
    Assumes each camera position belongs to exactly one edge, so devices never need to share raw images; location privacy is then claimed but models contain 3D point positions.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Radiant: Large-scale 3D Gaussian Rendering based on Hierarchical Framework." pith.science (2026). https://pith.science/paper/A5XAELIF

@misc{pith2026241205546,
  author       = {Pith},
  title        = {Pith review of: Radiant: Large-scale 3D Gaussian Rendering based on Hierarchical Framework},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/A5XAELIF}},
  note         = {Machine review of arXiv:2412.05546}
}
read the original abstract

With the advancement of computer vision, the recently emerged 3D Gaussian Splatting (3DGS) has increasingly become a popular scene reconstruction algorithm due to its outstanding performance. Distributed 3DGS can efficiently utilize edge devices to directly train on the collected images, thereby offloading computational demands and enhancing efficiency. However, traditional distributed frameworks often overlook computational and communication challenges in real-world environments, hindering large-scale deployment and potentially posing privacy risks. In this paper, we propose Radiant, a hierarchical 3DGS algorithm designed for large-scale scene reconstruction that considers system heterogeneity, enhancing the model performance and training efficiency. Via extensive empirical study, we find that it is crucial to partition the regions for each edge appropriately and allocate varying camera positions to each device for image collection and training. The core of Radiant is partitioning regions based on heterogeneous environment information and allocating workloads to each device accordingly. Furthermore, we provide a 3DGS model aggregation algorithm that enhances the quality and ensures the continuity of models' boundaries. Finally, we develop a testbed, and experiments demonstrate that Radiant improved reconstruction quality by up to 25.7\% and reduced up to 79.6\% end-to-end latency.

Figures

Figures reproduced from arXiv: 2412.05546 by the authors.

Figure 1
Figure 1. Left: Cloud-based 3DGS training. Middle: Cloud￾device-based 3DGS training. Right: Our hierarchical frame￾work for 3DGS training. 500×250m2 . Moreover, to obtain the initial model (i.e., the 3D Gaussian points) for the scene, all the raw images captured by the device must be uploaded to the cloud for structure-from￾motion [10] (SfM), not only increasing communication over￾head but also severely infringing on device p… view at source ↗
Figure 2
Figure 2. Heterogeneity when training across different scene [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 4
Figure 4. The performance (↑ PSNR, ↑ SSIM, ↓ LPIPS) of using model merge and model retrain algorithms in the same area. ground truth images followed by minimal retraining epochs of the entire model. As shown in [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figures from the paper (8 more)
Figure 5
Figure 5. Figure 5: The Overview of Radiant. (a) The Adaptive Region Planning algorithm iterates together with the Resource-aware Task [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: An example of Adaptive Region Planning algorithm. [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 7
Figure 7. Figure 7: An example of Resource-aware Task Partitioning [PITH_FULL_IMAGE:figures/full_fig_p006_7.png]
Figure 8
Figure 8. Figure 8: Model Aggregation. We concatenate device models [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]
Figure 9
Figure 9. Figure 9: Comparison of end-to-end latency. “even” means to [PITH_FULL_IMAGE:figures/full_fig_p008_9.png]
Figure 10
Figure 10. Figure 10: The partitions of Radiant and the latency of each [PITH_FULL_IMAGE:figures/full_fig_p008_10.png]
Figure 11
Figure 11. Figure 11: The impact of the number of aggregate epochs on the [PITH_FULL_IMAGE:figures/full_fig_p009_11.png]
Figure 13
Figure 13. Figure 13: The convergence process of the ARP algorithm. [PITH_FULL_IMAGE:figures/full_fig_p009_13.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references · 27 canonical work pages

  1. [8]

    Citygaussian: Real-time high-quality large-scale scene rendering with gaussians,

    Y . Liu, H. Guan, C. Luo, L. Fan, J. Peng, and Z. Zhang, “Citygaussian: Real-time high-quality large-scale scene rendering with gaussians,” in Proceedings of the European conference on computer vision (ECCV) ,

  2. [9]

    Dogaussian: Distributed-oriented gaussian splatting for large-scale 3d reconstruction via gaussian consensus,

    G. H. L. Yu Chen, “Dogaussian: Distributed-oriented gaussian splatting for large-scale 3d reconstruction via gaussian consensus,” inarXiv, 2024

  3. [1]

    3d gaussian splatting for real-time radiance field rendering.,

    B. Kerbl, G. Kopanas, T. Leimk ¨uhler, and G. Drettakis, “3d gaussian splatting for real-time radiance field rendering.,” ACM Trans. Graph. , vol. 42, no. 4, pp. 139–1, 2023

  4. [2]

    Recent advances in 3d gaussian splatting,

    T. Wu, Y .-J. Yuan, L.-X. Zhang, J. Yang, Y .-P. Cao, L.-Q. Yan, and L. Gao, “Recent advances in 3d gaussian splatting,” Computational Visual Media, pp. 1–30, 2024

  5. [3]

    Vdg: Vision-only dynamic gaussian for driving simulation,

    H. Li, J. Li, D. Zhang, C. Wu, J. Shi, C. Zhao, H. Feng, E. Ding, J. Wang, and J. Han, “Vdg: Vision-only dynamic gaussian for driving simulation,” arXiv preprint arXiv:2406.18198 , 2024

  6. [4]

    Darkgs: Learning neural illumination and 3d gaussians relighting for robotic exploration in the dark,

    T. Zhang, K. Huang, W. Zhi, and M. Johnson-Roberson, “Darkgs: Learning neural illumination and 3d gaussians relighting for robotic exploration in the dark,” arXiv preprint arXiv:2403.10814 , 2024

  7. [5]

    Scalability in perception for autonomous driving: Waymo open dataset,

    P. Sun, H. Kretzschmar, X. Dotiwalla, A. Chouard, V . Patnaik, P. Tsui, J. Guo, Y . Zhou, Y . Chai, B. Caine, et al. , “Scalability in perception for autonomous driving: Waymo open dataset,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pp. 2446–2454, 2020

  8. [6]

    Ggrt: Towards generalizable 3d gaussians without pose priors in real-time,

    H. Li, Y . Gao, D. Zhang, C. Wu, Y . Dai, C. Zhao, H. Feng, E. Ding, J. Wang, and J. Han, “Ggrt: Towards generalizable 3d gaussians without pose priors in real-time,” in Proceedings of the European conference on computer vision (ECCV) , 2024. Accepted

Show all 42 references
  1. [7]

    Vastgaussian: Vast 3d gaussians for large scene reconstruction,

    J. Lin, Z. Li, X. Tang, J. Liu, S. Liu, J. Liu, Y . Lu, X. Wu, S. Xu, Y . Yan, et al., “Vastgaussian: Vast 3d gaussians for large scene reconstruction,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 5166–5175, 2024

  2. [10]

    Photo tourism: exploring photo collections in 3d,

    N. Snavely, S. M. Seitz, and R. Szeliski, “Photo tourism: exploring photo collections in 3d,” in ACM siggraph 2006 papers , pp. 835–846, 2006

  3. [11]

    Fed3dgs: Scalable 3d gaussian splatting with federated learning,

    T. Suzuki, “Fed3dgs: Scalable 3d gaussian splatting with federated learning,” arXiv preprint arXiv:2403.11460 , 2024

  4. [12]

    3d gaussian splatting: Survey, technologies, challenges, and opportunities,

    Y . Bao, T. Ding, J. Huo, Y . Liu, Y . Li, W. Li, Y . Gao, and J. Luo, “3d gaussian splatting: Survey, technologies, challenges, and opportunities,” arXiv preprint arXiv:2407.17418 , 2024

  5. [13]

    Vector-indistinguishability: location dependency based privacy protection for successive location data,

    Y . Zhao and J. Chen, “Vector-indistinguishability: location dependency based privacy protection for successive location data,” IEEE Transac- tions on Computers , vol. 73, no. 4, pp. 970–979, 2023

  6. [14]

    Client-edge-cloud hierarchical federated learning,

    L. Liu, J. Zhang, S. Song, and K. B. Letaief, “Client-edge-cloud hierarchical federated learning,” in ICC 2020-2020 IEEE international conference on communications (ICC) , pp. 1–6, IEEE, 2020

  7. [15]

    Resource- efficient federated learning with hierarchical aggregation in edge com- puting,

    Z. Wang, H. Xu, J. Liu, H. Huang, C. Qiao, and Y . Zhao, “Resource- efficient federated learning with hierarchical aggregation in edge com- puting,” in IEEE INFOCOM 2021-IEEE conference on computer com- munications, pp. 1–10, IEEE, 2021

  8. [16]

    Optimizing training efficiency and cost of hierarchical federated learning in heterogeneous mobile- edge cloud computing,

    Y . Cui, K. Cao, J. Zhou, and T. Wei, “Optimizing training efficiency and cost of hierarchical federated learning in heterogeneous mobile- edge cloud computing,” IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 42, no. 5, pp. 1518–1531, 2022

  9. [17]

    Ewa splatting,

    M. Zwicker, H. Pfister, J. Van Baar, and M. Gross, “Ewa splatting,” IEEE Transactions on Visualization and Computer Graphics , vol. 8, no. 3, pp. 223–238, 2002

  10. [18]

    An in-depth study of uplink performance of 5g mmwave networks,

    M. Ghoshal, Z. J. Kong, Q. Xu, Z. Lu, S. Aggarwal, I. Khan, Y . Li, Y . C. Hu, and D. Koutsonikolas, “An in-depth study of uplink performance of 5g mmwave networks,” in Proceedings of the ACM SIGCOMM Workshop on 5G and Beyond Network Measurements, Modeling, and Use Cases, pp. ...

  11. [19]

    V oronoi diagrams—a survey of a fundamental geo- metric data structure,

    F. Aurenhammer, “V oronoi diagrams—a survey of a fundamental geo- metric data structure,” ACM Computing Surveys (CSUR), vol. 23, no. 3, pp. 345–405, 1991

  12. [20]

    Huawei cloud elastic cloud server (ecs),

    HUAWEI CLOUD, “Huawei cloud elastic cloud server (ecs),”

  13. [21]

    Image quality assessment: from error visibility to structural similarity,

    Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality assessment: from error visibility to structural similarity,” IEEE transactions on image processing , vol. 13, no. 4, pp. 600–612, 2004

  14. [22]

    The unreasonable effectiveness of deep features as a perceptual metric,

    R. Zhang, P. Isola, A. A. Efros, E. Shechtman, and O. Wang, “The unreasonable effectiveness of deep features as a perceptual metric,” in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 586–595, 2018

  15. [23]

    Mega-nerf: Scalable construction of large-scale nerfs for virtual fly-throughs,

    H. Turki, D. Ramanan, and M. Satyanarayanan, “Mega-nerf: Scalable construction of large-scale nerfs for virtual fly-throughs,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, pp. 12922–12931, 2022

  16. [24]

    3d gaussian splatting for real-time radiance field rendering,

    B. Kerbl, G. Kopanas, T. Leimk ¨uhler, and G. Drettakis, “3d gaussian splatting for real-time radiance field rendering,” ACM Transactions on Graphics, July 2023

  17. [25]

    Nerf: Representing scenes as neural radiance fields for view synthesis,

    B. Mildenhall, P. P. Srinivasan, M. Tancik, J. T. Barron, R. Ramamoorthi, and R. Ng, “Nerf: Representing scenes as neural radiance fields for view synthesis,” Communications of the ACM , vol. 65, no. 1, pp. 99–106, 2021

  18. [26]

    Efficient large-scale scene represen- tation with a hybrid of high-resolution grid and plane features,

    Y . Zhang, G. Chen, and S. Cui, “Efficient large-scale scene represen- tation with a hybrid of high-resolution grid and plane features,” arXiv preprint arXiv:2303.03003, 2023

  19. [27]

    Switch-nerf: Learning scene decomposition with mixture of experts for large-scale neural radiance fields,

    M. Zhenxing and D. Xu, “Switch-nerf: Learning scene decomposition with mixture of experts for large-scale neural radiance fields,” in The Eleventh International Conference on Learning Representations , 2022

  20. [28]

    Drone-nerf: Efficient nerf based 3d scene reconstruction for large-scale drone survey,

    Z. Jia, B. Wang, and C. Chen, “Drone-nerf: Efficient nerf based 3d scene reconstruction for large-scale drone survey,” Image and Vision Computing, vol. 143, p. 104920, 2024

  21. [29]

    Federated learning for large-scale scene modeling with neural radiance fields,

    T. Suzuki, “Federated learning for large-scale scene modeling with neural radiance fields,” arXiv preprint arXiv:2309.06030 , 2023

  22. [30]

    Building rome in a day,

    S. Agarwal, Y . Furukawa, N. Snavely, I. Simon, B. Curless, S. M. Seitz, and R. Szeliski, “Building rome in a day,” Communications of the ACM, vol. 54, no. 10, pp. 105–112, 2011

  23. [31]

    An automated method for large-scale, ground- based city model acquisition,

    C. Fr ¨uh and A. Zakhor, “An automated method for large-scale, ground- based city model acquisition,” International Journal of Computer Vision, vol. 60, pp. 5–24, 2004

  24. [32]

    Structure-from-motion revisited,

    J. L. Schonberger and J.-M. Frahm, “Structure-from-motion revisited,” in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 4104–4113, 2016

  25. [33]

    Block-nerf: Scalable large scene neural view synthesis,

    M. Tancik, V . Casser, X. Yan, S. Pradhan, B. Mildenhall, P. P. Srinivasan, J. T. Barron, and H. Kretzschmar, “Block-nerf: Scalable large scene neural view synthesis,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 8248–8258, 2022

  26. [34]

    Grid-guided neural radiance fields for large urban scenes,

    L. Xu, Y . Xiangli, S. Peng, X. Pan, N. Zhao, C. Theobalt, B. Dai, and D. Lin, “Grid-guided neural radiance fields for large urban scenes,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 8296–8306, 2023

  27. [35]

    Share: Shaping data distribution at edge for communication-efficient hierarchical federated learning,

    Y . Deng, F. Lyu, J. Ren, Y . Zhang, Y . Zhou, Y . Zhang, and Y . Yang, “Share: Shaping data distribution at edge for communication-efficient hierarchical federated learning,” in 2021 IEEE 41st International Con- ference on Distributed Computing Systems (ICDCS) , pp. 24–34, IEEE, 2021

  28. [36]

    Data heterogeneity-robust federated learning via group client selection in in- dustrial iot,

    Z. Li, Y . He, H. Yu, J. Kang, X. Li, Z. Xu, and D. Niyato, “Data heterogeneity-robust federated learning via group client selection in in- dustrial iot,” IEEE Internet of Things Journal, vol. 9, no. 18, pp. 17844– 17857, 2022

  29. [37]

    Optimal user-edge assignment in hierarchical federated learning based on statistical properties and network topology constraints,

    N. Mhaisen, A. A. Abdellatif, A. Mohamed, A. Erbad, and M. Guizani, “Optimal user-edge assignment in hierarchical federated learning based on statistical properties and network topology constraints,” IEEE Trans- actions on Network Science and Engineering , vol. 9, no. 1, pp. 5...

  30. [38]

    Flee: A hierarchical federated learning framework for distributed deep neural network over cloud, edge, and end device,

    Z. Zhong, W. Bao, J. Wang, X. Zhu, and X. Zhang, “Flee: A hierarchical federated learning framework for distributed deep neural network over cloud, edge, and end device,” ACM Transactions on Intelligent Systems and Technology (TIST), vol. 13, no. 5, pp. 1–24, 2022

  31. [39]

    A hierarchical knowledge transfer framework for heterogeneous federated learning,

    Y . Deng, J. Ren, C. Tang, F. Lyu, Y . Liu, and Y . Zhang, “A hierarchical knowledge transfer framework for heterogeneous federated learning,” in IEEE INFOCOM 2023-IEEE Conference on Computer Communications, pp. 1–10, IEEE, 2023

  32. [40]

    Hierarchical federated learning with momentum acceleration in multi-tier networks,

    Z. Yang, S. Fu, W. Bao, D. Yuan, and A. Y . Zomaya, “Hierarchical federated learning with momentum acceleration in multi-tier networks,” IEEE Transactions on Parallel and Distributed Systems , 2023

  33. [41]

    Taming subnet-drift in d2d- enabled fog learning: A hierarchical gradient tracking approach,

    E. Chen, S. Wang, and C. G. Brinton, “Taming subnet-drift in d2d- enabled fog learning: A hierarchical gradient tracking approach,” in IEEE INFOCOM 2024-IEEE conference on computer communications ,

  34. [2024]

    https://www.huaweicloud.com/intl/en-us/product/ecs.html, Last accessed on 2024-7-16

Pith tools

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