Timeliness-Fidelity Tradeoff in 3D Scene Representations
Not on the map. Why?
Abstract (arXiv)
Real-time three-dimensional (3D) scene representations serve as one of the building blocks that bolster various innovative applications, e.g., digital manufacturing, Virtual/Augmented/Extended/Mixed Reality (VR/AR/XR/MR), and the metaverse. Despite substantial efforts that have been made to real-time communications and computing, real-time 3D scene representations remain a challenging task. This paper investigates the tradeoff between timeliness and fidelity in real-time 3D scene representations. Specifically, we establish a framework to evaluate the impact of communication delay on the tradeoff, where the real-world scenario is monitored by multiple cameras that communicate with an edge server. To improve fidelity for 3D scene representations, we propose to use a single-step Proximal Policy Optimization (PPO) method that leverages the Age of Information (AoI) to decide if the received image needs to be involved in 3D scene representations and rendering. We test our framework and the proposed approach with different well-known 3D scene representation methods. Simulation results reveal that real-time 3D scene representation can be sensitively affected by communication delay, and our proposed method can achieve optimal 3D scene representation results.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 24 of its 36 references that are in the map. First posted on arXiv 23 Jul 2024.
Builds on
24 of 36 references.
References · 24
- 2017Proximal Policy Optimization Algorithmscited by 15,215
- 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
- 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,723
- 2015High-Dimensional Continuous Control Using Generalized Advantage Estimationcited by 2,026
- 2022TensoRF: Tensorial Radiance Fieldscited by 1,280
- 2023K-Planes: Explicit Radiance Fields in Space, Time, and Appearancecited by 746
- 2023Nerfstudio: A Modular Framework for Neural Radiance Field Developmentcited by 600
- 2021NICE-SLAM: Neural Implicit Scalable Encoding for SLAMcited by 589
- 2021iMAP: Implicit Mapping and Positioning in Real-Timecited by 476
- 2022MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstructioncited by 470
- 2021Advances in Neural Renderingcited by 365
- 2022NeRF-SLAM: Real-Time Dense Monocular SLAM with Neural Radiance Fieldscited by 188
12 more
- 2021Learning Task-Oriented Communication for Edge Inference: An Information Bottleneck Approachcited by 110
- 2022Personalized Saliency in Task-Oriented Semantic Communications: Image Transmission and Performance Analysiscited by 23
- 2020Unsupervised Learning of Depth, Optical Flow and Pose with Occlusion from 3D Geometrycited by 21
- 2022Computer Vision Aided mmWave Beam Alignment in V2X Communicationscited by 15
- 2020Single-step deep reinforcement learning for open-loop control of laminar and turbulent flowscited by 14
- 2023Neural radiance fields in the industrial and robotics domain: applications, research opportunities and use casescited by 10
- 2023NerfBridge: Bringing Real-time, Online Neural Radiance Field Training to Roboticscited by 9
- 2023Task-Oriented Cross-System Design for Timely and Accurate Modeling in the Metaversecited by 7
- 2023Effective Communication: When to Pull Updates?cited by 4
- 2017Status Updates Through Multicast Networkscited by 3
- 2020On the Age of Information in Internet of Things Systems with Correlated Devicescited by 3
- 2021Latency and Information Freshness in Multipath Communications for Virtual Realitycited by 2
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025Task-Oriented Communications for 3D Scene Representation: Balancing Timeliness and Fidelity
- 2025ExploreGS: Explorable 3D Scene Reconstruction with Virtual Camera Samplings and Diffusion Priors
- 20253D Scene Generation: A Survey
- 2026Beyond Pixel Histories: World Models with Persistent 3D State
- 2024Vid3D: Synthesis of Dynamic 3D Scenes using 2D Video Diffusion
- 2026Matrix-Game 3.5: Enhancing Real-Time Streaming Interactive World Models with Patch Memory
- 2026SceneAE: Scene Generation and Reconstruction in Queryable 3D-Anchored Latent Space
- 2025Voyager: Long-Range and World-Consistent Video Diffusion for Explorable 3D Scene Generation