Abstract
In conventional point cloud delivery, a sender uses octree-based digital video compression to stream three-dimensional (3D) points and the corresponding color attributes over band-limited links, e.g., wireless channels, for 3D scene reconstructions. However, the digital-based delivery schemes have an issue called cliff effect, where the 3D reconstruction quality is a step function in terms of wireless channel quality. We propose a novel scheme of point cloud delivery, called HoloCast, to gracefully improve the reconstruction quality with the improvement of wireless channel quality. HoloCast regards the 3D points and color components as graph signals and directly transmits linear-transformed signals based on graph Fourier transform (GFT), without digital quantization and entropy coding operations. One of main contributions in HoloCast is that the use of GFT can deal with non-ordered and non-uniformly distributed multi-dimensional signals such as holographic data unlike conventional delivery schemes. Performance results with point cloud data show that HoloCast yields better 3D reconstruction quality compared to digital-based methods in noisy wireless environment.
Introduction
Streaming 3D point clouds—the data structures behind holographic displays and AR/VR—is incredibly bandwidth-intensive. A single uncompressed point cloud with 800,000 points can require nearly 38 Mbits. Current delivery methods rely on digital compression (like octree-based PCL), which suffer from two major flaws in wireless environments:
- The Cliff Effect: Digital bitstreams are fragile. If the wireless channel quality (SNR) drops below a certain threshold, the decoder fails to synchronize, and the 3D scene fails to reconstruct entirely.
- The Leveling Effect: Because digital systems use fixed quantization, the reconstruction quality hits a ceiling. Even if the wireless signal becomes perfect, the quality won't improve beyond the limits of the initial compression.
Furthermore, point clouds are "irregular"—unlike 2D images, points are not on a neat grid and are unordered. This makes standard compression transforms (like the Discrete Cosine Transform) difficult to apply effectively.
Methods: The HoloCast Architecture
The authors propose HoloCast, which shifts the paradigm from digital bits to "soft" analog-like transmission. Instead of standard video transforms, HoloCast uses the Graph Fourier Transform (GFT). This mathematical tool "compacts" the energy of the 3D coordinates and color attributes into a few spectral coefficients, even for non-uniformly distributed data.
As shown in the pipeline above, the system follows these key steps:
- Graph Representation: The point cloud is represented as a weighted, undirected graph where edges are drawn between nearby points using a Gaussian kernel to capture spatial correlations.
- No Quantization: Crucially, HoloCast skips the quantization and entropy coding steps that cause the cliff effect.
- Power Allocation & Analog Mapping: GFT coefficients are scaled based on their power to minimize Mean Squared Error (MSE) and then mapped directly to the In-phase (I) and Quadrature (Q) components of the wireless signal.
- MMSE Decoding: At the receiver, a Minimum Mean-Square Error filter suppresses noise before performing an inverse GFT to reconstruct the 3D scene.
Key Results
The researchers evaluated HoloCast using reference point clouds like pencil_10_0 and office1. The results demonstrate that HoloCast achieves "graceful degradation," where reconstruction quality improves linearly with wireless signal strength.
As illustrated in the performance graph, digital schemes (QPSK, 16QAM, 256QAM) exhibit a sharp "cliff" where performance drops to zero below specific SNR thresholds. In contrast, HoloCast (GFT) maintains a linear relationship between SNR and reconstruction quality (sMSE).
Key Findings include: * Massive Gains: In noisy environments, HoloCast achieved up to a 39.45 dB improvement in coordinate accuracy and a 32.48 dB improvement in color accuracy compared to standard digital PCL-based delivery. * GFT Superiority: The use of GFT exploits correlations in non-ordered data much more effectively than DCT or systems with no decorrelation. * Scalability: The benefits of HoloCast actually increased with the number of points, showing better relative reconstruction quality on larger datasets.
Conclusion
HoloCast successfully addresses the fragility of digital point cloud streaming by treating 3D data as signals on a graph. By replacing rigid digital compression with GFT-based soft transmission, it ensures that users with better wireless connections receive higher-quality holographic experiences without the risk of total signal loss in challenging environments. Future work will focus on reducing the computational complexity of the GFT eigen-decomposition for massive point clouds.
Summary updated