Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

A quantum secure direct communication link carrying learned semantic codes for 3D point clouds reaches an equivalent data rate above the Wyner secrecy capacity, with a 46.3-fold gain at code length 10.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-04 06:48 UTC pith:352QWAV3

load-bearing objection The 'beyond Shannon-Wyner capacity' claim is an accounting artifact—EDR credits reconstructed bits—but the integrated QSDC point-cloud demo is real. the 3 major comments →

arxiv 2511.07760 v1 pith:352QWAV3 submitted 2025-11-11 quant-ph

Quantum Semantic Communication Beyond the Shannon-Wyner Channel Capacity

Min Wang , Gui-Fa Zhu , Guo-Fei Long , Jianxing Guo , Yu-Chen Liu , Dong Pan , Li-Ping Nong , Gui-Lu Long This is my paper
classification quant-ph PACS 03.67.Hk
keywords quantum secure direct communicationsemantic communication3D point cloudspoint cloud compressionequivalent data rateWyner secrecy capacityChamfer distanceSTIKE protocol
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper sets out to show that two known weaknesses cancel out: quantum secure direct communication is secure but slow, while semantic communication is fast but insecure. It trains a neural codec that maps a 2048-point 3D point cloud into a short semantic code (length 10 to 300 numbers), transmits that code over a 50 km QSDC link, and reconstructs the full cloud at the receiver. The experiment reports an equivalent data rate up to 1591 kbps at code length 10, a 46.3-fold gain over direct transmission, and at length 50 a rate that surpasses the Wyner secrecy capacity with low distortion. The paper interprets these numbers as beating channel-capacity limits, and as removing the main obstacle to large-scale secure point-cloud applications. The key metric, equivalent data rate, counts reconstructed bits rather than transmitted bits; that accounting convention is the premise on which the capacity claim rests.

Core claim

On its own terms, the paper demonstrates that a semantic codec and a QSDC transport can be trained and run as one end-to-end system, shifting the bottleneck from the channel to the codec's fidelity. The authors report a QBER of 3.31% and a raw communication rate of 37.36 kbps on a 50 km standard single-mode fiber link. With the codec, the equivalent data rate is 1591.52 kbps at code length 10, 731.44 kbps at length 50, and 263.05 kbps at length 300, versus 34.37 kbps for raw point-cloud transmission. They claim that at n = 50 the rate exceeds the Wyner secrecy capacity of 560.20 kbps, and at n = 10 it even exceeds the channel's mutual information of 1496.53 kbps, with Chamfer Distance droppi

What carries the argument

The engine is the learned semantic codec paired with the STIKE one-way QSDC protocol. The encoder builds a local graph via k-nearest neighbors, passes it through five high-dimensional feature extraction layers (graph convolutional), max-pools to a 1×n code, and power-normalizes; the decoder uses transposed-convolution point-restoration layers plus point-conditioned feature-restoration layers to expand the code back to N×3. The quantum channel transmits the code under STIKE, which uses one-time-pad encryption before photon transmission, decoy states to detect eavesdropping, and LDPC plus spread-spectrum coding. Training is end-to-end by minimizing Chamfer Distance. The metrics EDR (equivalent

Load-bearing premise

The central claim rests on the paper's Equivalent Data Rate definition, which counts all reconstructed point-cloud bits as delivered; counting only the bits actually sent over the channel (about 2.6 kbps at code length 10) removes the capacity-beating result.

What would settle it

Recompute EDR from transmitted bits rather than reconstructed bits. For n=10, transmitted data is 10×32 bits per cloud; with the reported total times, the rate is about 2.6 kbps, below the measured raw channel rate of 37.36 kbps. Making this substitution shows that neither the Wyner secrecy capacity nor the Shannon mutual information is exceeded, which would refute the central beyond-capacity claim.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Share X Bluesky LinkedIn Reddit HN

If this is right

  • If EDR is accepted as a performance measure, a single 50 km QSDC link can carry 3D point-cloud tasks at effective rates above the secrecy capacity, making bandwidth-limited secure applications plausible.
  • At code length 50 the system keeps a 21.3-fold efficiency gain with Chamfer Distance 2.0e-3, so moderate code lengths offer a practical quality-versus-speed trade-off.
  • Because encode and decode times (a few milliseconds) are far below channel time (about 500 ms), future improvements in raw QSDC rate would translate almost one-to-one into EDR gains.
  • With QBER at 3.31%, below the STIKE threshold, the security guarantee of eavesdropping detection appears to survive when semantic coding is inserted into the transmission chain.

Where Pith is reading between the lines

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

  • The capacity comparison is only as strong as the EDR convention; counting only the n×32 bits that physically cross the link would make the n=10 rate about 2.6 kbps, far below the raw 37.36 kbps channel rate, so the 'beyond capacity' statement follows from reconstructed-bit accounting rather than from channel-rate physics.
  • The same codec design could be transferred from object point clouds to other structured data such as terrain scans, vehicle LiDAR frames, or medical geometry, where learned priors exist, and tested on the same link.
  • A direct test of robustness would be to vary fiber distance or link loss while holding n fixed and measure how Chamfer Distance degrades; this would separate semantic distortion from channel-induced distortion.
  • One could also compare against a classical encrypted channel carrying the same codec and the same lossy reconstruction; that would cleanly separate the security contribution from the compression contribution.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The paper proposes a quantum semantic communication scheme for 3D point clouds, combining a neural semantic autoencoder (encoder/decoder networks) with a one-way QSDC link over 50 km of standard single-mode fiber. The authors report an Equivalent Data Rate (EDR) of up to 1591.52 kbps for a semantic code length n=10, a 46.30-fold Relative Transmission Efficiency over direct transmission, and claim that this 'surpasses both Wyner and Shannon capacity limits.' The experimental part includes a quasi-QSDC system with measured QBER 3.31% and a raw channel rate of 37.36 kbps, plus reconstruction quality evaluated by Chamfer Distance.

Significance. If the headline claim were supported, this would be a striking result: a practical demonstration of semantic communication over QSDC exceeding fundamental information-theoretic limits. The paper does contain a real experimental QSDC link with detailed hardware parameters, and it systematically varies the semantic code length n, reporting both reconstruction distortion (CD) and timing data. However, the central claim is not supported. As argued below, the EDR metric credits the receiver with all reconstructed point-cloud bits, not the bits actually transmitted through the quantum channel. The comparison to Shannon-Wyner capacity is therefore an accounting artifact rather than a channel result. The potential value of the work lies in a more modest claim: a semantic compression scheme for point clouds operating over a QSDC link.

major comments (3)
  1. [RESULTS, Table I and Fig. 3(e)] The load-bearing claim 'EDR exceeds Shannon-Wyner capacity' rests on the definition of EDR as 'the ratio between the total bits of all reconstructed point clouds and the total transmission time.' This counts N×3×32 = 196,608 bits per point cloud at the receiver, while only n×32 bits were actually transmitted. For n=10, the transmitted payload is 320 bits per point cloud. With the reported total transmission time of 1,244,715 ms, the transmitted-bit rate is approximately 2.6 kbps—not 1591.52 kbps. This is far below the measured QSDC channel rate (37.36 kbps) and below both quoted capacity lines (560.20 and 1496.53 kbps). The 'beyond capacity' statement is thus forced by the EDR normalization, not by the experiment.
  2. [RESULTS, Fig. 3(e) caption] The two reference capacity values (560.20 kbps and 1496.53 kbps) are asserted without derivation. No formula, protocol-specific calculation, or numerical inputs are provided to show how the 'secrecy capacity derived from Shannon-Wyner entropy' and the 'mutual information between the quantum transmitter and quantum receiver' are obtained from the stated system parameters (photon number, loss, detector efficiency, dark count, QBER, etc.). Since the entire paper revolves around comparing EDR to these values, the absence of a checkable derivation makes the central comparison unverifiable.
  3. [RESULTS, Fig. 3(d)-(f); DISCUSSION] The EDR metric conflates source coding with channel transmission. The decoder uses a learned prior on the ShapeNet dataset to reconstruct a dense point cloud from a short semantic code; the reconstructed bits are not independent information conveyed through the quantum channel. Even if EDR were redefined to count only transmitted bits, the semantic code is a lossy source code with nonzero distortion (CD≈2×10^-3 at n=50). Comparing its output bit count directly to Shannon or Wyner capacity is conceptually inappropriate: the correct statement is that semantic coding reduces the required bitrate for a given distortion level (a source-coding gain), not that the system transmits information above the channel capacity. The paper's conclusion 'surpassing both Wyner and Shannon capacity limits' is therefore unsupported.
minor comments (4)
  1. [Title] The title contains a spacing error: 'C hannel' should read 'Channel'.
  2. [Transmission Data Analysis] There is a duplicated phrase: 'The results are reported in The results are reported in' should be corrected.
  3. [Fig. 3(b) caption] The term 'Numbers' is unclear; it appears to refer to batches or transmission rounds. Please use a consistent and descriptive label.
  4. [Experiment and Results] The term 'quasi-QSDC' is used without definition. Clarify whether it is synonymous with the STIKE one-way QSDC protocol or denotes a specific practical variant.

Circularity Check

1 steps flagged

The claim that EDR exceeds Shannon-Wyner capacity is forced by defining EDR as reconstructed bits per second while capacities are channel bits per second.

specific steps
  1. self definitional [Results, 'Equivalent Data Rate' definition and Fig. 3(e) comparison; Table I]
    "EDR is defined as the ratio between the total bits of all reconstructed point clouds and the total transmission time under a given code length n ... the EDR at n = 50 already exceeds the Shannon-Wyner channel capacity with a CD value of 2.00 × 10−3, and the EDR at n = 10 even exceeds the mutual information between the quantum transmitter and quantum receiver at the expense of data quality."

    The quoted Shannon and Wyner capacities are rates of channel-transmitted bits per second. EDR instead counts reconstructed decoder outputs (N×3 coordinates per point cloud) per second, even though only n semantic features are transmitted. With n=10 and N=2048, the output is 2048×3 coordinates per cloud compared with n transmitted features, so EDR exceeds the transmitted-bit rate by a factor of about 614.4 (before protocol overhead). Thus any lossy codec with n<6144 will 'exceed capacity' under this metric purely by construction, regardless of what the quantum channel can actually carry. The 46.30-fold RTE time-saving claim is a separate and legitimate metric, but the 'surpassing Shannon-Wyner capacity' conclusion is determined by the definition of EDR rather than by an information-theoreti

full rationale

The paper's central numerical claim is not a measured channel rate; it arises from replacing transmitted bits with reconstructed bits in the rate definition. This is the load-bearing step and makes the beyond-capacity conclusion true by definition. The transmission-time savings, CD values, and RTE are experimentally grounded and not circular. The authors do cite several of their own QSDC works, including the same-month TechRxiv reference [17] for the concept of quantum semantic communication, but the capacity comparison does not rest on those citations, so I do not treat those as load-bearing. The circularity is an accounting/metric-definition issue rather than a hidden fit or self-citation chain.

Axiom & Free-Parameter Ledger

2 free parameters · 4 axioms · 0 invented entities

The system depends on three classes of load-bearing choices: (i) a hand-selected code length n per trained model; (ii) the codec weights, which are fitted to ShapeNet and determine the actual compression gain and distortion; (iii) the metric convention that reconstructed bits may be counted as if transmitted. The capacity reference values are quoted as external protocol results. No new physical entities are introduced.

free parameters (2)
  • semantic code length n = 10, 20, 50, 100, 200, 300
    Hand-selected hyperparameter controlling compression; a separate model is trained per n, and EDR/RTE depend directly on n.
  • neural codec parameters (encoder E and decoder D) = not disclosed (trained on ShapeNet)
    The compression gain and reconstruction quality are properties of the trained network; the 46.30-fold gain is measured for these fitted weights, not derived.
axioms (4)
  • domain assumption The STIKE quasi-QSDC protocol provides one-time-pad-encrypted transmission while simultaneously distilling enough fresh key to sustain the OTP at the measured 37.36 kbps payload rate.
    Invoked in the Experimental section; no key-balance or security analysis is given for the demonstrated payload rate.
  • ad hoc to paper The EDR metric — counting all 6144×32 bits of the reconstructed point cloud as transmitted data — is the correct unit for comparing against channel capacity.
    Introduced in Results and used for the 'beyond capacity' claim; without this convention the claim collapses.
  • ad hoc to paper Reconstruction at CD ≈ 2e-3 (n=50) is treated as equivalent to lossless transmission for bit-counting purposes.
    EDR credits full reconstruction bits even though the codec is lossy and introduces visible distortion (Fig. 3f).
  • standard math Standard capacity formulas (Shannon / Wyner wiretap) apply to the quoted QSDC capacity numbers.
    The values 560.20 and 1496.53 kbps are quoted without derivation; the paper relies on the standard theory from [10,13].

pith-pipeline@v1.3.0-alltime-deepseek · 8353 in / 17454 out tokens · 154882 ms · 2026-08-04T06:48:06.921541+00:00 · methodology

0 comments
Cite this review

Pith. "Pith review of Quantum Semantic Communication Beyond the Shannon-Wyner Channel Capacity." pith.science (2026). https://pith.science/paper/352QWAV3

@misc{pith2026251107760,
  author       = {Pith},
  title        = {Pith review of: Quantum Semantic Communication Beyond the Shannon-Wyner Channel Capacity},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/352QWAV3}},
  note         = {Machine review of arXiv:2511.07760}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Quantum Secure Direct Communication (QSDC), a paradigm-shifting breakthrough in quantum communication, exploits quantum states for unmediated information transmission. Rooted in the inviolable fundamental laws of quantum mechanics, QSDC enables ultrasensitive detection of even the faintest eavesdropping attempts, guaranteeing true communication security solely when no interference exists. If eavesdropping or intrusion is detected mid-transmission, the system instantly alerts users and severs data flow, shielding them from unauthorized tracking and mitigating hacker threats. Over two decades, QSDC has seen extraordinary advancements, currently attaining kilobit-per-second transmission over 100 km of commercial optical fiber. However, its practical scalability remains constrained by insufficient transmission rates, a critical bottleneck. Semantic communication, which drastically boosts transmission efficiency by extracting core information features, nevertheless stays vulnerable to malicious intrusions. Integrating these paradigms promises to simultaneously enhance equivalent data rate and security. Herein, we propose and experimentally validate a quantum semantic communication scheme, applying it to 3D point clouds. It achieves a 46.30-fold efficiency gain over direct transmission, surpassing both Wyner and Shannon capacity limits. This breakthrough not only clears the path for large-scale QSDC deployment but also marks a pivotal milestone in quantum information science.

Figures

Figures reproduced from arXiv: 2511.07760 by Dong Pan, Gui-Fa Zhu, Gui-Lu Long, Guo-Fei Long, Jianxing Guo, Li-Ping Nong, Min Wang, Yu-Chen Liu.

Figure 1
Figure 1. Figure 1: FIG. 1: General framework for quantum semantic [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2: Framework of the quantum semantic [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3: (a) Communication rate and QBER versus time of the QSDC sy [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

discussion (0)

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

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. A Comprehensive Survey on Semantic Communication in Non-Terrestrial Networks: Architectures, Methodologies, and Challenges

    cs.IT 2026-05 unverdicted novelty 2.0

    A literature survey that pairs NTN limitations with semantic communication properties, organizes work by platform and methodology, and lists open problems for integrated SAGIN systems.

Reference graph

Works this paper leans on

34 extracted references · 1 canonical work pages · cited by 1 Pith paper

  1. [1]

    At the transmitter, the semantic encoder processes ∗ These authors contributed equally to this work

    The communica- tion process begins with the information source, which generates raw data (including point clouds, speech, and text). At the transmitter, the semantic encoder processes ∗ These authors contributed equally to this work. † nongliping@gxnu.edu.cn ‡ gllong@tsinghua.edu.cn the input data by leveraging the source semantic knowl- edge base: this k...

    Pith/arXiv arXiv 2025

  2. [2]

    (1) where ai and bj denote the i-th and j-th points in P and P′, respectively, with i, j = 1, 2, . . . , N. RESULTS We used the ShapeNet[ 32] dataset in our experiments. The dataset contains 51,127 point cloud samples across 55 common object categories, such as airplanes, chairs, and beds. It was divided into 35,708 training samples, 5,158 validation samp...

    2048

  3. [3]

    Theoretically efficient high- capacity quantum-key-distribution scheme

    Long, G.-L., Liu, X.S. Theoretically efficient high- capacity quantum-key-distribution scheme. Phys. Rev. A 65, 032302 (2002)

    2002

  4. [4]

    X., Lu, J., and Hanzo, L

    Pan, D., Long, G.-L., Yin, L., Sheng, Y.-B., Ruan, D., Ng, S. X., Lu, J., and Hanzo, L. The Evolution of Quan- tum Secure Direct Communication: On the Road to the Qinternet. IEEE Communications Surveys & Tutorials 26(3), 1898-1949 (2024)

    1949

  5. [5]

    Y., Yu, B., Jing, M

    Hu, J. Y., Yu, B., Jing, M. Y. et al. Experimental quan- tum secure direct communication with single photons. Light Sci. Appl. 5, 16144 (2016)

    2016

  6. [6]

    S., Sheng, Y

    Zhang, W., Ding, D. S., Sheng, Y. B. et al. Quantum Secure Direct Communication with Quantum Memory. Phys. Rev. Lett. 118, 220501 (2017)

    2017

  7. [7]

    B., Huang, Y

    Zhu, F., Zhang, W., Sheng, Y. B., Huang, Y. D. Experi- mental long-distance quantum secure direct communica- tion. Sci. Bull. 62, 1519 (2017)

    2017

  8. [8]

    Qi, R., Sun, Z., Lin, Z. et al. Implementation and security analysis of practical quantum secure direct communica- tion. Light Sci. Appl. 8, 22 (2019)

    2019

  9. [9]

    Qi, Z.D., Li, Y.H., Huang, Y.W. et al. A 15-user quantum secure direct communication network. Light Sci. Appl. 10, 183 (2021)

    2021

  10. [10]

    R., Sun, Z., Qi, R

    Zhang, H. R., Sun, Z., Qi, R. Y. et al. Realization of quantum secure direct communication over 100 km fiber with time-bin and phase quantum states. Light Sci. Appl. 11, 1-9 (2022)

    2022

  11. [11]

    Wang, M., Zhang, W., Guo, J. et al. Experimental Demonstration of Secure Relay in Quantum Secure Di- rect Communication Network. Entropy 25, 1548 (2023)

    2023

  12. [12]

    Simul- taneous transmission of information and key exchange us- ing the same photonic quantum states

    Pan, D., Liu, Y.-C., Niu, P., Zhang, H., Zhang, F., Wang, M., Song, X.-T., Chen, X., Zheng, C., Long, G.-L. Simul- taneous transmission of information and key exchange us- ing the same photonic quantum states. Science Advances 11(8), eadt4627 (2025)

    2025

  13. [13]

    A 300-km fully-connected quantum secure direct com- munication network

    Yang, Y., Li, Y., Li, H., Wu, C., Zheng, Y., Chen, X. A 300-km fully-connected quantum secure direct com- munication network. Science Bulletin 70(9), 1445-1451 (2025)

    2025

  14. [14]

    Quantum secure direct commu- nication: whispering with photons

    Wang, M., Long, G-L. Quantum secure direct commu- nication: whispering with photons. National Science Re- view 12(8), nwaf096 (2025)

    2025

  15. [15]

    Wyner, A. D. The wire-tap channel. Bell Syst. Tech. J. 54, 1355–1387 (1975)

    1975

  16. [16]

    Z. Qin, X. Tao, J. Lu, W. Tong, and G. Y. Li, Se- mantic communications: Principles and challenges, arXiv preprint arXiv:2201.01389 (2021)

    Pith/arXiv arXiv 2021

  17. [17]

    H. Xie, Z. Qin, G. Y. Li, and B.-H. Juang, Deep learning enabled semantic communication systems, IEEE Trans- actions on Signal Processing 69, 2663–2675 (2021)

    2021

  18. [18]

    W. Yang, H. Du, Z. Q. Liew, W. Y. B. Lim, Z. Xiong, D. Niyato, X. Chi, X. Shen, and C. Miao, Semantic com- munications for future internet: Fundamentals, applica- tions, and challenges, IEEE Communications Surveys & Tutorials 25(1), 213–250 (2022)

    2022

  19. [19]

    Pan and G

    D. Pan and G. L. Long, Quantum Semantic Com- munication. TechRxiv. October 27, 2025. DOI: 10.36227/techrxiv.176153499.91712883/v1 (2025)

    arXiv 2025

  20. [20]

    Lattice-based access authentica- tion scheme for quantum communication networks

    Wang, M., Long, G.-L. Lattice-based access authentica- tion scheme for quantum communication networks. Sci. China Inform. Sci. 67(12), 222501 (2024)

    2024

  21. [21]

    Airborne laser scanning — an intro- duction and overview

    Wehr, A., Lohr, U. Airborne laser scanning — an intro- duction and overview. ISPRS Journal of Photogrammetry and Remote Sensing 54(2-3), 68-82 (1999)

    1999

  22. [22]

    R., Su, H., Mo, K., and Guibas, L

    Qi, C. R., Su, H., Mo, K., and Guibas, L. J. Point- Net: Deep learning on point sets for 3D classification and segmentation. In Proceedings of the IEEE Confer- ence on Computer Vision and Pattern Recognition , 652- 660 (2017)

    2017

  23. [23]

    D., and Schindler, K

    Hackel, T., Wegner, J. D., and Schindler, K. Fast seman- tic segmentation of 3D point clouds with strongly varying density. ISPRS Annals 3(3), 177-184 (2016)

    2016

  24. [24]

    Task-oriented scene graph-based semantic communications with adaptive channel coding

    Sun, S., Qin, Z., Xie, H., Tao, X. Task-oriented scene graph-based semantic communications with adaptive channel coding. IEEE Transactions on Wireless Com- munications, (2024)

    2024

  25. [25]

    Synchronous semantic communications for video and speech

    Tian, Y., Ying, J., Qin, Z., Jin, Y., Tao, X. Synchronous semantic communications for video and speech. In: ICC 2024 - IEEE International Conference on Communica- tions, pp. 3396–3401. IEEE, (2024)

    2024

  26. [26]

    Large-language-model-enabled text semantic commu- nication systems

    Wang, Z., Zou, L., Wei, S., Li, K., Liao, F., Mi, H., Lai, R. Large-language-model-enabled text semantic commu- nication systems. Applied Sciences 15(13), 7227 (2025). https://doi.org/10.3390/app15137227

    work page doi:10.3390/app15137227 2025

  27. [27]

    Semantic Communication of Point Clouds Based on Deep Neural Networks

    Chen, D., et al. Semantic Communication of Point Clouds Based on Deep Neural Networks. IEEE Trans- actions on Communications 70(6), 3958-3972 (2022)

    2022

  28. [28]

    Quantum cryptography

    Gisin, N., Ribordy, G., Tittel, W., and Zbinden, H. Quantum cryptography. Reviews of Modern Physics 74(1), 145 (2002)

    2002

  29. [29]

    Nearest neighbor pattern classifica- tion

    Cover, T., & Hart, P. Nearest neighbor pattern classifica- tion. IEEE Transactions on Information Theory 13(1), 21-27 (1967)

    1967

  30. [30]

    Kipf, T. N. Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907 (2016)

    Pith/arXiv arXiv 2016

  31. [31]

    Wireless point cloud transmission

    Bian, C., Shao, Y., & G¨ und¨ uz, D. Wireless point cloud transmission. In 2024 IEEE 25th International Workshop 7 on Signal Processing Advances in Wireless Communica- tions (SPA WC), 851-855 (2024)

    2024

  32. [32]

    Point cloud pre-training with diffusion models

    Zheng, X., Huang, X., Mei, G., Hou, Y., Lyu, Z., Dai, B., Ouyang, W., & Gong, Y. Point cloud pre-training with diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, 22935-22945 (2024)

    2024

  33. [33]

    Fan, H., Su, H., & Guibas, L. J. A point set generation network for 3D object reconstruction from a single im- age. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 605-613 (2017)

    2017

  34. [34]

    X., Funkhouser, T., Guibas, L., Hanrahan, P., Huang, Q., Li, Z., Savarese, S., Savva, M., Song, S., Su, H., Xiao, J., Yi, L., Yu, F

    Chang, A. X., Funkhouser, T., Guibas, L., Hanrahan, P., Huang, Q., Li, Z., Savarese, S., Savva, M., Song, S., Su, H., Xiao, J., Yi, L., Yu, F. ShapeNet: An Information-Rich 3D Model Repository. arXiv preprint arXiv:1512.03012 (2015)

    Pith/arXiv arXiv 2015