pith. sign in

arxiv: 2605.24290 · v1 · pith:4JADSP3Knew · submitted 2026-05-22 · 💻 cs.NI

RxGS: Receiver-Generalizable 3D Gaussian Splatting for Radio-Frequency Data Synthesis

Kang Yang , Mani Srivastava This is my paper

Pith reviewed 2026-06-30 13:57 UTC · model grok-4.3

classification 💻 cs.NI
keywords 3D Gaussian SplattingRF data synthesisreceiver generalizationwireless signal predictionneural renderingscene representationmulti-receiver rendering
0
0 comments X

The pith

RxGS uses one shared 3D Gaussian Splatting model to synthesize RF signals for any receiver position in a scene.

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

RxGS introduces a receiver-generalizable method for predicting radio-frequency signals at arbitrary transmitter and receiver positions. It splits the task into two stages: first learning the scene geometry that does not depend on the receiver, then learning directional radiance that does depend on receiver location while keeping the geometry fixed. Global and local conditioning branches capture shared and per-scatterer receiver effects, and a multi-receiver rasterizer enables efficient batch rendering. A sympathetic reader would care because the approach replaces N separate models with one, supports unseen receivers, and reduces training cost by up to 45 times, inference cost by 7.6 times, and storage by a factor of N.

Core claim

RxGS achieves receiver-generalizable synthesis within a single unified model. A first stage learns shared 3D Gaussian geometry, and a second stage freezes it and learns directional radiance conditioned on receiver position. A global conditioning branch captures shared receiver-dependent effects across the scene, while a local branch models per-scatterer variations from the receiver's geometry and occlusion. A multi-receiver CUDA rasterizer batches rendering across all N receivers. Evaluated across various RF datasets, RxGS matches or improves over per-receiver baselines with a single shared model and generalizes to receivers unseen during training within the scene.

What carries the argument

Two-stage 3D Gaussian Splatting pipeline that first learns receiver-independent scene geometry then freezes those Gaussians to learn receiver-position-conditioned directional radiance via global and local branches.

If this is right

  • One model replaces N independent models for N receivers in the same scene.
  • Signal prediction works at receiver positions never seen during training.
  • Training cost drops by up to 45 times and inference cost by 7.6 times.
  • Storage scales with one model instead of scaling linearly with the number of receivers.
  • Batch rendering of multiple receivers becomes possible through the multi-receiver CUDA rasterizer.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The geometry-versus-radiance split could transfer to acoustic or optical synthesis tasks where listener or viewer position varies but scene structure does not.
  • Real-time wireless planning tools might become practical once the reduced per-receiver overhead is combined with existing ray-tracing accelerators.
  • Extending the local branch to handle moving scatterers could allow the same model to track changes without full retraining.

Load-bearing premise

Scene geometry stays the same regardless of receiver location while only the directional radiance changes with receiver position.

What would settle it

Train the model on a subset of receivers in a scene and test on a new receiver position; if accuracy falls substantially below that of a model trained specifically for the new receiver, the generalization claim is falsified.

Figures

Figures reproduced from arXiv: 2605.24290 by Kang Yang, Mani Srivastava.

Figure 1
Figure 1. Figure 1: RxGS’s two-stage architecture. Stage I learns receiver-independent 3D Gaussian geometry, and Stage II conditions radiance on the receiver via global and local modules in one unified model. 4.1 Geometry-Radiance Decomposition Workflow The key insight behind RxGS is that the 3D Gaussian attributes partition into two groups with distinct dependencies. The geometric attributes, position pk, covariance Ck, and … view at source ↗
Figure 2
Figure 2. Figure 2: Unseen-receiver generalization. “Ours” is [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Localization: RxGS bridges 71% of the sparse-to-dense gap. Fingerprint-based indoor localization trades off accuracy against gateway deployment cost. Dense deployments are accurate but expensive, while sparse deployments are cheap but imprecise. We show that RxGS bridges this gap by syn￾thesizing fingerprints at virtual gateway positions. Follow￾ing the protocol of §5.2, we assume only N = 5 gateways are p… view at source ↗
Figure 4
Figure 4. Figure 4: Floor plan of the conference room (8 m × 6 m × 3 m) used for the RFID spectrum dataset. F.1 BLE RSSI The BLE RSSI dataset is collected in a real-world nursing home occupying ∼15,000 ft2 [5]. NRX = 21 BLE gateways operating at 2.4 GHz are deployed throughout the facility, each measuring the received signal strength (RSSI in dBm) from a moving BLE node. TX positions are collected by random walks using 30 BLE… view at source ↗
Figure 5
Figure 5. Figure 5: Per-receiver MAE breakdown on BLE RSSI for each baseline and its receiver-conditioned [PITH_FULL_IMAGE:figures/full_fig_p023_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Per-receiver PSNR breakdown on RFID spatial spectrum for each baseline and its receiver [PITH_FULL_IMAGE:figures/full_fig_p024_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Per-receiver SNR breakdown on WiFi CSI for each baseline and its receiver-conditioned [PITH_FULL_IMAGE:figures/full_fig_p025_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Qualitative CSI traces on the WiFi CSI dataset at three transmitter (TX) positions. Top [PITH_FULL_IMAGE:figures/full_fig_p026_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Qualitative comparison of RFID spatial spectrum prediction. [PITH_FULL_IMAGE:figures/full_fig_p028_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Scaling with the number of training receivers. Seen accuracy stays flat while unseen [PITH_FULL_IMAGE:figures/full_fig_p029_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Sensitivity of RxGS to the Stage I reference receiver choice on BLE. Each of the first 21 bars corresponds to using one gateway as the reference, and the avg bar uses the averaged signal across all receivers. All single-reference variants fall within 0.5 dBm of the best, and the averaged￾signal initialization achieves the lowest per-receiver MAE. 10 20 30 40 X (m) 10 20 30 40 50 60 Y (m) Survey-Only (Gap … view at source ↗
Figure 12
Figure 12. Figure 12: Coverage heatmaps for K = 5 access-point deployment on the BLE dataset. Each cell shows the fraction of test transmitters in a ∼ 1.6 × 1.6 m region whose strongest-AP RSSI exceeds −80 dBm; gray cells contain no measurements. SURVEY-ONLY deploys at 5 surveyed gateways, OURS uses RxGS predictions to select 5 from all 21 candidates, and UPPER BOUND selects the optimal 5 from full-survey RSSI. OURS cuts the g… view at source ↗
read the original abstract

Radio-frequency (RF) data synthesis predicts the received signal given transmitter and receiver positions, and is essential for wireless applications. Recent 3D Gaussian Splatting (3DGS)-based methods achieve efficient synthesis at any transmitter but only for a fixed receiver. Therefore, supporting $N$ receivers in one scene requires $N$ independent models and precludes prediction at unseen receivers. We present RxGS, which achieves receiver-generalizable synthesis within a single unified model. Our key insight is that scene geometry is receiver-independent while directional radiance is not: a first stage learns shared 3D Gaussian geometry, and a second stage freezes it and learns directional radiance conditioned on receiver position. A global conditioning branch captures shared receiver-dependent effects across the scene, while a local branch models per-scatterer variations from the receiver's geometry and occlusion. A multi-receiver CUDA rasterizer further batches rendering across all $N$ receivers. Evaluated across various RF datasets, RxGS matches or improves over per-receiver baselines with a single shared model and generalizes to receivers unseen during training within the scene, cutting training cost by up to $45\times$, inference cost by $7.6\times$, and storage by $N\times$.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 3 minor

Summary. The paper introduces RxGS, a two-stage 3D Gaussian Splatting approach for RF data synthesis that learns a single shared model for multiple receivers. Stage 1 learns receiver-independent scene geometry via 3D Gaussians; stage 2 freezes those Gaussians and learns receiver-conditioned directional radiance using global and local conditioning branches. A custom multi-receiver CUDA rasterizer supports batched rendering. The central claims are that RxGS matches or exceeds per-receiver baselines, generalizes to unseen receivers within the same scene, and yields up to 45× training, 7.6× inference, and N× storage savings.

Significance. If the empirical claims hold, the work offers a practical route to scalable RF synthesis for multi-receiver and dynamic-receiver wireless applications by exploiting the geometry-radiance separation. The explicit two-stage freezing and the global/local conditioning design are clear engineering contributions that directly address the storage and retraining costs of prior per-receiver 3DGS methods.

minor comments (3)
  1. [Abstract, §1] Abstract and §1 state performance parity and generalization but supply no quantitative tables, error bars, dataset sizes, or ablation details; the full experimental section should include these to allow verification of the 45×/7.6× claims.
  2. [Method (rasterizer subsection)] The multi-receiver CUDA rasterizer is described at a high level; pseudocode or a short complexity analysis would clarify how batching across N receivers is implemented without introducing per-receiver overhead.
  3. [§3.2] Notation for the global and local conditioning branches (e.g., how receiver position is encoded and injected) should be made fully explicit with equations to aid reproducibility.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive summary, recognition of the geometry-radiance separation and conditioning design as engineering contributions, and the recommendation of minor revision. No major comments were provided in the report.

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The provided abstract and context describe a two-stage modeling approach grounded in an explicit premise (receiver-independent geometry vs. receiver-dependent radiance). This premise directly motivates the freezing step and conditioning branches without any equations, fitted parameters renamed as predictions, or self-citation chains. No load-bearing derivation reduces to its own inputs by construction; the method is a consistent implementation of the stated decomposition. The central claims about generalization and cost savings are empirical outcomes, not tautological re-statements of inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract supplies only the high-level architectural split; no explicit free parameters, axioms, or invented entities are named.

axioms (1)
  • domain assumption Scene geometry is receiver-independent while directional radiance is receiver-dependent
    This premise is invoked to justify the two-stage training procedure.

pith-pipeline@v0.9.1-grok · 5744 in / 1226 out tokens · 32912 ms · 2026-06-30T13:57:21.347218+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

52 extracted references · 5 canonical work pages · 2 internal anchors

  1. [1]

    WiFi Sensing with Channel State Information: A Survey

    Yongsen Ma, Gang Zhou, and Shuangquan Wang. WiFi Sensing with Channel State Information: A Survey. ACM Computing Surveys, 52(3):1–36, 2019

    2019

  2. [2]

    WiTAG: Seamless WiFi Backscater Communication

    Ali Abedi, Farzan Dehbashi, Mohammad Hossein Mazaheri, Omid Abari, and Tim Brecht. WiTAG: Seamless WiFi Backscater Communication. InACM Special Interest Group on Data Communications (SIG- COMM), 2020

    2020

  3. [3]

    Rappaport.Wireless Communications: Principles and Practice

    Theodore S. Rappaport.Wireless Communications: Principles and Practice. Prentice Hall, 2002

    2002

  4. [4]

    Integrated Two-Way Radar Backscatter Communication and Sensing with Low-Power IoT Tags

    Ryu Okubo, Luke Jacobs, Jinhua Wang, Steven Bowers, and Elahe Soltanaghai. Integrated Two-Way Radar Backscatter Communication and Sensing with Low-Power IoT Tags. InACM Special Interest Group on Data Communications (SIGCOMM), 2024

    2024

  5. [5]

    NeRF 2: Neural Radio-Frequency Radiance Fields

    Xiaopeng Zhao, Zhenlin An, Qingrui Pan, and Lei Yang. NeRF 2: Neural Radio-Frequency Radiance Fields. InACM Annual International Conference on Mobile Computing and Networking (MobiCom), 2023

    2023

  6. [6]

    GeRaF: Neural Geometry Reconstruction from Radio Frequency Signals

    Jiachen Lu, Hailan Shanbhag, and Haitham Al Hassanieh. GeRaF: Neural Geometry Reconstruction from Radio Frequency Signals. InConference on Neural Information Processing Systems (NeurIPS), 2025

    2025

  7. [7]

    Can NeRFs See without Cameras?arXiv preprint arXiv:2505.22441, 2025

    Chaitanya Amballa, Sattwik Basu, Yu-Lin Wei, Zhijian Yang, Mehmet Ergezer, and Romit Roy Choudhury. Can NeRFs See without Cameras?arXiv preprint arXiv:2505.22441, 2025

    work page arXiv 2025

  8. [8]

    Diffusion Model-based RSSI Fingerprint Generation for Indoor Localization in Dynamic Environments

    Liuyi Yang, Patrick Finnerty, and Chikara Ohta. Diffusion Model-based RSSI Fingerprint Generation for Indoor Localization in Dynamic Environments. InIEEE Wireless and Optical Communications Conference (WOCC), pages 402–407, 2025

    2025

  9. [9]

    FLog: Automated modeling of link quality for LoRa networks in orchards.ACM Transactions on Sensor Networks, 21(2):1–28, 2025

    Kang Yang, Yuning Chen, and Wan Du. FLog: Automated modeling of link quality for LoRa networks in orchards.ACM Transactions on Sensor Networks, 21(2):1–28, 2025

    2025

  10. [10]

    Link Quality Modeling for LoRa Networks in Orchards

    Kang Yang, Yuning Chen, Tingruixiang Su, and Wan Du. Link Quality Modeling for LoRa Networks in Orchards. InACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN), 2023

    2023

  11. [11]

    OrchLoc: In-Orchard Localization via a Single LoRa Gateway and Generative Diffusion Model-Based Fingerprinting

    Kang Yang, Yuning Chen, and Wan Du. OrchLoc: In-Orchard Localization via a Single LoRa Gateway and Generative Diffusion Model-Based Fingerprinting. InACM International Conference on Mobile Systems, Applications, and Services (MobiSys), 2024

    2024

  12. [12]

    Generative Diffusion Model-Assisted Efficient Fingerprinting for In-Orchard Localization.IEEE Transactions on Mobile Computing, 24(12):12833–12851, 2025

    Kang Yang, Yuning Chen, and Wan Du. Generative Diffusion Model-Assisted Efficient Fingerprinting for In-Orchard Localization.IEEE Transactions on Mobile Computing, 24(12):12833–12851, 2025

    2025

  13. [13]

    Site Survey and Radio Frequency Planning for the Deployment of Next Generation WLAN

    Pushpendu Kar and Bhasker Dappuri. Site Survey and Radio Frequency Planning for the Deployment of Next Generation WLAN. InTopical Conference on Wireless Sensors and Sensor Networks (WiSNet), 2018

    2018

  14. [14]

    Understand Site Survey Guidelines for WLAN Deployment

    Cisco. Understand Site Survey Guidelines for WLAN Deployment. https://www. cisco.com/c/en/us/support/docs/wireless/5500-series-wireless-controllers/ 116057-site-survey-guidelines-wlan-00.html, 2023. [Online]

    2023

  15. [15]

    MRI: Model-Based Radio Interpolation for Indoor War-Walking.IEEE Transactions on Mobile Computing, 14(6):1231–1244, 2014

    Hyojeong Shin, Yohan Chon, Yungeun Kim, and Hojung Cha. MRI: Model-Based Radio Interpolation for Indoor War-Walking.IEEE Transactions on Mobile Computing, 14(6):1231–1244, 2014

    2014

  16. [16]

    Aranda, José A

    Felipe Parralejo, Fernando J. Aranda, José A. Paredes, Fernando J. Álvarez, and Jorge Morera. Comparative Study of Different BLE Fingerprint Reconstruction Techniques. InInternational Conference on Indoor Positioning and Indoor Navigation (IPIN), 2021. 10

    2021

  17. [17]

    Ray Tracing for Radio Propagation Modeling: Principles and Applications.IEEE Access, 3:1089–1100, 2015

    Zhengqing Yun and Magdy F Iskander. Ray Tracing for Radio Propagation Modeling: Principles and Applications.IEEE Access, 3:1089–1100, 2015

    2015

  18. [18]

    Opal: An Open Source Ray-Tracing Propagation Simulator for Electromagnetic Characterization.Plos One, 16(11):e0260060, 2021

    Esteban Egea-Lopez, Jose Maria Molina-Garcia-Pardo, Martine Lienard, and Pierre Degauque. Opal: An Open Source Ray-Tracing Propagation Simulator for Electromagnetic Characterization.Plos One, 16(11):e0260060, 2021

    2021

  19. [19]

    Indoor MIMO-OFDM Communication Link Using Ray Tracing

    MATLAB. Indoor MIMO-OFDM Communication Link Using Ray Tracing. https://www.mathworks. com/help/comm/ug/indoor-mimo-ofdm-communication-link-using-ray-tracing.html ,

  20. [20]

    Rad-NeRF: Ray-Decoupled Training of Neural Radiance Field.Conference on Neural Information Processing Systems (NeurIPS), 37:113742–113771, 2024

    Lidong Guo, Xuefei Ning, Yonggan Fu, Tianchen Zhao, Zhuoliang Kang, Jincheng Yu, Yingyan Celine Lin, and Yu Wang. Rad-NeRF: Ray-Decoupled Training of Neural Radiance Field.Conference on Neural Information Processing Systems (NeurIPS), 37:113742–113771, 2024

    2024

  21. [21]

    RF-3DGS: Wireless Channel Modeling with Radio Radiance Field and 3D Gaussian Splatting.IEEE Journal on Selected Areas in Communications, 2024

    Lihao Zhang, Haijian Sun, Samuel Berweger, Camillo Gentile, and Rose Qingyang Hu. RF-3DGS: Wireless Channel Modeling with Radio Radiance Field and 3D Gaussian Splatting.IEEE Journal on Selected Areas in Communications, 2024

    2024

  22. [22]

    WRF-GS+: Wireless Radiation Field Reconstruction with 3D Gaussian Splatting.IEEE Transactions on Wireless Communications, 2025

    Chaozheng Wen, Jingwen Tong, Yingdong Hu, Zehong Lin, and Jun Zhang. WRF-GS+: Wireless Radiation Field Reconstruction with 3D Gaussian Splatting.IEEE Transactions on Wireless Communications, 2025

    2025

  23. [23]

    GSRF: Complex-Valued 3D Gaussian Splatting for Efficient Radio-Frequency Data Synthesis.arXiv preprint arXiv:2502.01826, 2025

    Kang Yang, Gaofeng Dong, Sijie Ji, Wan Du, and Mani Srivastava. GSRF: Complex-Valued 3D Gaussian Splatting for Efficient Radio-Frequency Data Synthesis.arXiv preprint arXiv:2502.01826, 2025

    work page arXiv 2025

  24. [24]

    NeWRF: A Deep Learn- ing Framework for Wireless Radiation Field Reconstruction and Channel Prediction

    Haofan Lu, Christopher Vattheuer, Baharan Mirzasoleiman, and Omid Abari. NeWRF: A Deep Learn- ing Framework for Wireless Radiation Field Reconstruction and Channel Prediction. InInternational Conference on Machine Learning (ICML), 2024

    2024

  25. [25]

    3D Gaussian Splatting for Real-Time Radiance Field Rendering

    Bernhard Kerbl, Georgios Kopanas, Thomas Leimkühler, and George Drettakis. 3D Gaussian Splatting for Real-Time Radiance Field Rendering. InACM International Conference on Computer Graphics and Interactive Techniques (SIGGRAPH), 2023

    2023

  26. [26]

    Mip-Splatting: Alias-Free 3D Gaussian Splatting

    Zehao Yu, Anpei Chen, Binbin Huang, Torsten Sattler, and Andreas Geiger. Mip-Splatting: Alias-Free 3D Gaussian Splatting. InIEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024

    2024

  27. [27]

    A Survey on 3D Gaussian Splatting

    Guikun Chen and Wenguan Wang. A Survey on 3D Gaussian Splatting.arXiv preprint arXiv:2401.03890, 2024

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  28. [28]

    Massive MIMO Systems for 5G and beyond Networks—overview, Recent Trends, Challenges, and Future Research Direction.Sensors, 20(10):2753, 2020

    Robin Chataut and Robert Akl. Massive MIMO Systems for 5G and beyond Networks—overview, Recent Trends, Challenges, and Future Research Direction.Sensors, 20(10):2753, 2020

    2020

  29. [29]

    Cell-Free Massive Multiple-Input Multiple-Output Challenges and Opportunities: A Survey.ICT Express, 10(1):194–212, 2024

    Mahnoor Ajmal, Ayesha Siddiqa, Bomi Jeong, Junho Seo, and Dongkyun Kim. Cell-Free Massive Multiple-Input Multiple-Output Challenges and Opportunities: A Survey.ICT Express, 10(1):194–212, 2024

    2024

  30. [30]

    Spherical Harmonics.Computer Graphics and Multimedia Group, 18, 2005

    V olker Schönefeld. Spherical Harmonics.Computer Graphics and Multimedia Group, 18, 2005

    2005

  31. [31]

    Spherical Wave Expansion with Arbitrary Origin for Near-Field Antenna Measurements.IEEE Transactions on Antennas and Propagation, 65(8):4385–4388, 2017

    Rasmus Cornelius and Dirk Heberling. Spherical Wave Expansion with Arbitrary Origin for Near-Field Antenna Measurements.IEEE Transactions on Antennas and Propagation, 65(8):4385–4388, 2017

    2017

  32. [32]

    Generalizable Radio-Frequency Radiance Fields for Spatial Spectrum Synthesis

    Kang Yang, Yuning Chen, and Wan Du. Generalizable Radio-Frequency Radiance Fields for Spatial Spectrum Synthesis.arXiv preprint arXiv:2502.05708, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  33. [33]

    NeRF-APT: A New NeRF Framework for Wireless Channel Prediction

    Jingzhou Shen, Tianya Zhao, Yanzhao Wu, and Xuyu Wang. NeRF-APT: A New NeRF Framework for Wireless Channel Prediction. InIEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), 2025

    2025

  34. [34]

    V oxelRF: V oxelized Radiance Field for Fast Wireless Channel Modeling.IEEE Communications Letters, 30:617–621, 2025

    Zihang Zeng, Shu Sun, Meixia Tao, Yin Xu, and Xianghao Yu. V oxelRF: V oxelized Radiance Field for Fast Wireless Channel Modeling.IEEE Communications Letters, 30:617–621, 2025

    2025

  35. [35]

    NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis.Communications of the ACM, 65(1):99–106, 2021

    Ben Mildenhall, Pratul P Srinivasan, Matthew Tancik, Jonathan T Barron, Ravi Ramamoorthi, and Ren Ng. NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis.Communications of the ACM, 65(1):99–106, 2021

    2021

  36. [36]

    MVS- NeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereo

    Anpei Chen, Zexiang Xu, Fuqiang Zhao, Xiaoshuai Zhang, Fanbo Xiang, Jingyi Yu, and Hao Su. MVS- NeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereo. InIEEE/CVF Interna- tional Conference on Computer Vision (ICCV), 2021. 11

    2021

  37. [37]

    Deformable 2D Gaussian Splatting for Efficient Wireless Radiance Field Rendering.IEEE Transactions on Visualization and Computer Graphics, pages 1–17, 2026

    Mufan Liu, Cixiao Zhang, Qi Yang, Yujie Cao, Yiling Xu, Yin Xu, Shu Sun, Mingzeng Dai, and Yunfeng Guan. Deformable 2D Gaussian Splatting for Efficient Wireless Radiance Field Rendering.IEEE Transactions on Visualization and Computer Graphics, pages 1–17, 2026

    2026

  38. [38]

    Scalable 3D Gaussian Splatting-Based RF Signal Spatial Propagation Modeling

    Kang Yang, Wan Du, and Mani Srivastava. Scalable 3D Gaussian Splatting-Based RF Signal Spatial Propagation Modeling. InACM Conference on Embedded Networked Sensor Systems (SenSys), 2025

    2025

  39. [39]

    NeRD: Neural Reflectance Decomposition from Image Collections

    Mark Boss, Raphael Braun, Varun Jampani, Jonathan T Barron, Ce Liu, and Hendrik Lensch. NeRD: Neural Reflectance Decomposition from Image Collections. InIEEE/CVF International Conference on Computer Vision (ICCV), 2021

    2021

  40. [40]

    NeRFactor: Neural Factorization of Shape and Reflectance under an Unknown Illumination.ACM Transactions on Graphics, 40(6):1–18, 2021

    Xiuming Zhang, Pratul P Srinivasan, Boyang Deng, Paul Debevec, William T Freeman, and Jonathan T Barron. NeRFactor: Neural Factorization of Shape and Reflectance under an Unknown Illumination.ACM Transactions on Graphics, 40(6):1–18, 2021

    2021

  41. [41]

    Relightable 3D Gaussians: Realistic Point Cloud Relighting with Brdf Decomposition and Ray Tracing

    Jian Gao, Chun Gu, Youtian Lin, Zhihao Li, Hao Zhu, Xun Cao, Li Zhang, and Yao Yao. Relightable 3D Gaussians: Realistic Point Cloud Relighting with Brdf Decomposition and Ray Tracing. InEuropean Conference on Computer Vision (ECCV), 2024

    2024

  42. [42]

    GS-IR: 3D Gaussian Splatting for Inverse Rendering

    Zhihao Liang, Qi Zhang, Ying Feng, Ying Shan, and Kui Jia. GS-IR: 3D Gaussian Splatting for Inverse Rendering. InIEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024

    2024

  43. [43]

    Relightable Gaussian Codec Avatars

    Shunsuke Saito, Gabriel Schwartz, Tomas Simon, Junxuan Li, and Giljoo Nam. Relightable Gaussian Codec Avatars. InIEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024

    2024

  44. [44]

    Sionna: An open-source library for next-generation physical layer research,

    Jakob Hoydis, Sebastian Cammerer, Fayçal Ait Aoudia, Avinash Vem, Nikolaus Binder, Guillermo Marcus, and Alexander Keller. Sionna: An Open-Source Library for Next-Generation Physical Layer Research. arXiv preprint arXiv:2203.11854, 2022

    work page arXiv 2022

  45. [45]

    Sionna RT: Differentiable Ray Tracing for Radio Propagation Modeling

    Jakob Hoydis, Fayçal Aït Aoudia, Sebastian Cammerer, Merlin Nimier-David, Nikolaus Binder, Guillermo Marcus, and Alexander Keller. Sionna RT: Differentiable Ray Tracing for Radio Propagation Modeling. In IEEE Globecom Workshops (GC Wkshps), 2023

    2023

  46. [46]

    Ultra Dense Indoor MaMIMO CSI Dataset

    Sibren De Bast and Sofie Pollin. Ultra Dense Indoor MaMIMO CSI Dataset. https://dx.doi.org/10. 21227/nr6k-8r78, 2025. [Online]

    2025

  47. [47]

    Image Quality Metrics: PSNR Vs

    A Hore and D Ziou. Image Quality Metrics: PSNR Vs. SSIM. International Conference on Pattern Recognition. InInternational Conference on Pattern Recognition (ICPR), 2010

    2010

  48. [48]

    Understanding Real Many-Antenna MU- MIMO Channels

    Clayton Shepard, Jian Ding, Ryan E Guerra, and Lin Zhong. Understanding Real Many-Antenna MU- MIMO Channels. InAsilomar Conference on Signals, Systems and Computers (ACSCC), 2016. Appendix Contents A Proof of Geometry-Radiance Decomposition 14 B Proof of Last-Segment Factorization 16 C Proofs of Corollaries 1–3 17 C.1 Proof of Corollary 1 . . . . . . . . ...

    2016

  49. [49]

    The geometric attributes{p k,C k, τk}are receiver-independent. 14

  50. [50]

    The radiance ϕk =ϕ k ˆdk,r is receiver-dependent, with receiver-gradient given by Equation(27)under the per-scatterer identification

  51. [51]

    we were unable to find the license for the dataset we used

    Under the per-scatterer identification, the radiance is the only term in Equation(21) that carries receiver-dependence; all remaining factors are receiver-independent. Proof. Step 1 (Per-scatterer decomposition).Each propagation path terminates at the receiver, with its final segment originating from a scatterer sl,Pl for Pl ≥1 , or directly from t for th...

    2040

  52. [52]

    Guidelines: • The answer [N/A] means that the paper does not involve crowdsourcing nor research with human subjects

    Institutional review board (IRB) approvals or equivalent for research with human subjects Question: Does the paper describe potential risks incurred by study participants, whether such risks were disclosed to the subjects, and whether Institutional Review Board (IRB) approvals (or an equivalent approval/review based on the requirements of your country or ...