Latency in Real-Time 3D Volumetric Streaming: A Comprehensive Study

Hosun Yoon; Inayat Ali; Seong Moon; Seungwoo Hong

arxiv: 2605.22131 · v2 · pith:DMIGKPMJnew · submitted 2026-05-21 · 💻 cs.NI

Latency in Real-Time 3D Volumetric Streaming: A Comprehensive Study

Seungwoo Hong , Hosun Yoon , Seong Moon , Inayat Ali This is my paper

Pith reviewed 2026-05-22 02:56 UTC · model grok-4.3

classification 💻 cs.NI

keywords latency measurement3D volumetric streamingreal-time streamingnetwork layertransport protocolapplication layerVR AR latencybottleneck identification

0 comments

The pith

Breaking real-time 3D volumetric streaming into three layers identifies the main sources of latency and points to fixes.

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 measure and explain delays in systems that stream high-fidelity 3D models for VR, AR, gaming, and telepresence. It divides the end-to-end process into an application layer, a transport-protocol layer, and a network layer, then runs tests inside an actual running system. The measurements show how much each layer contributes to total delay and what inside each layer creates the wait. From those numbers the authors derive concrete changes that reduce the waits. A reader would care because lower latency directly improves immersion and cuts motion sickness in interactive 3D environments.

Core claim

Evaluating each layer in a real-world volumetric streaming system reveals the latency bottlenecks, quantifies their relative impact, and identifies the underlying causes of delay, which in turn supports targeted optimization strategies that reduce overall latency and improve responsiveness.

What carries the argument

The three-layer breakdown of the streaming pipeline (application layer, transport protocol layer, network layer) together with direct measurement of each layer inside a live system.

If this is right

Optimization efforts can be concentrated on the layer that contributes the largest measured share of delay.
Changes to application-layer processing or protocol settings can produce measurable reductions in end-to-end latency.
The same measurement method can be reused to check whether proposed fixes actually lower the identified bottlenecks.
Best-practice guidelines for building responsive volumetric systems can be written directly from the quantified layer contributions.

Where Pith is reading between the lines

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

The same layer-wise measurement approach could be applied to other real-time 3D or volumetric media such as point-cloud telepresence.
Adding automated, continuous layer monitoring to deployed systems would let operators react to latency spikes as they occur.
Combining the reported optimizations with edge-server placement might shrink the network-layer component further.

Load-bearing premise

The real-world test system and the way delays were recorded accurately capture the delays that would appear in ordinary deployments.

What would settle it

Running the same layer-by-layer timing tests in a second, independent deployment and finding that the relative sizes of the three latency components differ sharply from the reported values.

Figures

Figures reproduced from arXiv: 2605.22131 by Hosun Yoon, Inayat Ali, Seong Moon, Seungwoo Hong.

**Figure 2.** Figure 2: Experiment setup and configuration The authors in [9] focus on point cloud compression, improving online compression efficiency by reducing spatial and temporal redundancies in continuous point clouds. They utilize deep learning-based compression methods to minimize temporal redundancies and store 3D cloud data in 2D matrices, achieving better performance than traditional methods like Octree and MPEG. The… view at source ↗

**Figure 4.** Figure 4: 3D streaming structure for volumetric live conferencing [PITH_FULL_IMAGE:figures/full_fig_p003_4.png] view at source ↗

**Figure 5.** Figure 5: Hierarchical overview of the 3D streaming system architecture [PITH_FULL_IMAGE:figures/full_fig_p003_5.png] view at source ↗

**Figure 7.** Figure 7: illustrates the method used to measure detailed performance metrics. To measure frame latency, a timestamp was added to the data structure of the transmission frame. A sending timestamp was inserted when transmitting the first segment of the frame, and when the receiver receives the first segment, the difference between the received time and the timestamp is calculated to determine the network latency. Fra… view at source ↗

**Figure 6.** Figure 6: Layered latency measurement model of individual IP packets. Therefore, performance measurement of latency can be categorized into three layers: Application Layer Service and Frame Latency, Protocol Layer Nodeto-Node Frame Latency, and Network Layer Node-to-Node Packet Latency. IV. MEASUREMENT RESULT AND EVALUATION This section describes measurement methods, metrics, and results for the three layers. The m… view at source ↗

**Figure 9.** Figure 9: Service latency, frame latency, and sync server distribution time [PITH_FULL_IMAGE:figures/full_fig_p004_9.png] view at source ↗

**Figure 11.** Figure 11: presents the average values of the measured protocol latency. The latency in the left figure (Sender to Sync Server) was measured at 15.5ms, while the latency in the right figure (Sync Server to Receiver) was 19.8ms. The difference in latencies results from setting different transmission rates from the sender to server and server to the receiver which are 2Gbps and 1.5Gbps respectively. This demonstrates … view at source ↗

**Figure 12.** Figure 12: Protocol latency of sender to server and server to receiver [PITH_FULL_IMAGE:figures/full_fig_p005_12.png] view at source ↗

**Figure 13.** Figure 13: Network latency of sender to sync server to receiver [PITH_FULL_IMAGE:figures/full_fig_p005_13.png] view at source ↗

read the original abstract

Real-time 3D volumetric streaming is a transformative technology that enables the seamless transmission and rendering of high-fidelity 3D models, enhancing applications in virtual reality (VR), augmented reality (AR), gaming, telepresence, and remote collaboration. However, latency remains a major challenge, affecting immersion, causing motion sickness, and disrupting real-time interactions. Addressing these latency issues is essential for improving user experience and ensuring system efficiency. This study conducts a comprehensive latency measurement and analysis within a real-time volumetric streaming environment. We systematically break down the streaming process into three key layers: the application layer, the transport protocol layer, and the network layer. By evaluating each layer in a real-world system, we identify latency bottlenecks, quantify their impact, and uncover the underlying causes of delay. Based on these findings, we propose targeted optimization strategies to mitigate latency and enhance system responsiveness. Through this research, we establish best practices and innovative solutions to improve the efficiency, scalability, and overall user experience of real-time 3D volumetric streaming. Our insights contribute to advancing the field, paving the way for more immersive and responsive digital environments.

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.

Desk Editor's Note private letter to a colleague

This is a basic empirical breakdown of latency in volumetric streaming with layer analysis and optimization ideas, but thin on numbers and validation.

read the letter

The main thing to know is that the paper measures latency in a real-time 3D volumetric streaming system by splitting the process into application, transport protocol, and network layers, then flags bottlenecks and offers targeted fixes based on those measurements. It focuses on practical issues in VR, AR, and telepresence setups where delays hurt immersion and cause problems like motion sickness. The layer-by-layer approach is a standard way to isolate sources of delay, and applying it here to volumetric data gives a concrete view of how each part contributes in an actual deployment. That structure makes the suggestions for optimizations feel grounded in the observed delays rather than abstract. The real-world evaluation is a plus compared to simulation-only work, as it can surface issues like protocol overhead or network variability that matter in practice. On the downside, the available description gives no specific delay values, error bars, dataset details, or statistical checks, which makes it hard to judge whether the bottlenecks are accurately quantified or if the causes are correctly pinned down. The setup's representativeness of typical conditions also needs checking to rule out measurement artifacts. This work fits for networking and multimedia researchers who handle low-latency streaming for immersive apps. Someone building or tuning similar systems could get usable ideas from the decomposition, even if the methods are extensions of existing techniques rather than new frameworks. The central measurement logic holds up without obvious internal contradictions, so the paper deserves peer review to let referees examine the full experiments and data.

Referee Report

2 major / 2 minor

Summary. The manuscript claims to conduct a comprehensive latency measurement and analysis in a real-time 3D volumetric streaming environment. It systematically breaks down the streaming process into application, transport protocol, and network layers; evaluates each layer in a real-world system to identify bottlenecks, quantify their impact, and uncover underlying causes of delay; and proposes targeted optimization strategies to mitigate latency and enhance system responsiveness, with the goal of establishing best practices for VR, AR, gaming, telepresence, and remote collaboration applications.

Significance. If the empirical results are robust, representative, and free of instrumentation artifacts, the layer-by-layer breakdown could provide actionable insights for reducing end-to-end latency in volumetric streaming systems. The identification of specific bottlenecks and the derivation of mitigation strategies from quantified measurements would strengthen practical guidance in the field and support more immersive real-time 3D experiences.

major comments (2)

[Abstract] Abstract: the abstract outlines the approach and claims to quantify impact and uncover causes but supplies no quantitative results, error bars, dataset descriptions, statistical validation, or even summary statistics, so it is impossible to determine whether the measurements support the stated claims about bottlenecks and causes.
[Evaluation] Evaluation / Measurement Setup (assumed section describing the real-world system): the description must demonstrate that the chosen system and instrumentation are representative of typical deployment conditions and that recorded delays accurately reflect all sources without significant artifacts or unaccounted variables; otherwise the per-layer contributions and proposed optimizations rest on an unverified foundation.

minor comments (2)

[Throughout] Ensure all acronyms (VR, AR, etc.) are defined at first use and that figure captions explicitly state the measurement conditions and units for latency values.
[Related Work] Add a dedicated related-work subsection that positions the layer-wise breakdown against prior latency studies in volumetric or point-cloud streaming.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback on our manuscript. We have revised the paper to strengthen the abstract with quantitative details and to enhance the evaluation section's discussion of representativeness and measurement validity. Point-by-point responses follow.

read point-by-point responses

Referee: [Abstract] Abstract: the abstract outlines the approach and claims to quantify impact and uncover causes but supplies no quantitative results, error bars, dataset descriptions, statistical validation, or even summary statistics, so it is impossible to determine whether the measurements support the stated claims about bottlenecks and causes.

Authors: We agree that the abstract would benefit from including key quantitative results to better substantiate the claims. In the revised manuscript, we have updated the abstract to incorporate summary statistics drawn from the evaluation, including average per-layer latency contributions, dataset size (number of streaming sessions), and notes on statistical measures such as standard deviations. This revision makes the empirical support for the identified bottlenecks and optimization strategies more transparent to readers. revision: yes
Referee: [Evaluation] Evaluation / Measurement Setup (assumed section describing the real-world system): the description must demonstrate that the chosen system and instrumentation are representative of typical deployment conditions and that recorded delays accurately reflect all sources without significant artifacts or unaccounted variables; otherwise the per-layer contributions and proposed optimizations rest on an unverified foundation.

Authors: We appreciate this observation. The Evaluation section of the manuscript already specifies the real-world testbed, including commodity hardware and network conditions representative of VR/AR and telepresence deployments, along with layer-specific instrumentation using high-resolution timestamps. To further address the concern, we have expanded the section with additional details on validation procedures, including cross-checks for instrumentation overhead and confirmation that all primary delay sources (encoding, queuing, transmission, and rendering) are captured. These additions reinforce that the measurements form a reliable basis for the reported bottlenecks and optimizations. revision: yes

Circularity Check

0 steps flagged

Empirical measurement study with no derivation chain

full rationale

The paper conducts a real-world latency measurement study by breaking down volumetric streaming into application, transport, and network layers, quantifying bottlenecks from system evaluation, and proposing optimizations directly from those measurements. No equations, predictions, first-principles derivations, or self-citations are presented that reduce to fitted inputs or prior author results by construction. The central claims rest on external system data and direct observation rather than internal redefinition or statistical forcing, rendering the work self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The abstract describes an empirical measurement study and does not introduce or rely on explicit free parameters, mathematical axioms, or new postulated entities.

pith-pipeline@v0.9.0 · 5733 in / 1035 out tokens · 39642 ms · 2026-05-22T02:56:08.683265+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/AbsoluteFloorClosure.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

We systematically break down the streaming process into three key layers: the application layer, the transport protocol layer, and the network layer. By evaluating each layer in a real-world system, we identify latency bottlenecks...

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.