Search papers, labs, and topics across Lattice.
This paper introduces a robust framework for 3D Gaussian Splatting (3DGS) that addresses the fragility of bitstreams under packet loss during network streaming. By employing anchor-level atomic packaging and stratified random grouping, the method mitigates the impact of lost packets, transforming corrupted attributes into manageable missing-anchor erasures. The proposed approach significantly enhances rendering quality, achieving only about 3 dB average PSNR degradation under 20% random packet loss compared to the lossless HAC++ reference.
Packet loss can turn high-fidelity neural rendering into a visual nightmare, but this new framework keeps the quality intact by smartly packaging and recovering 3D attributes.
3D Gaussian Splatting (3DGS) and recent compression schemes such as HAC++ enable high-fidelity real-time neural rendering, but their bitstreams are fragile under packet loss during network streaming. Existing compression methods often separate correlated anchor attributes into independent streams, so losing one packet can create attribute-inconsistent broken anchors and severe rendering artifacts. We propose a packet-loss robust 3DGS transmission and error concealment framework. On the encoder side, anchor-level atomic packaging jointly encapsulates all attributes of each anchor, converting corrupted-attribute failures into clean missing-anchor erasures. Stratified random grouping further disperses packet losses across the spatial domain to avoid large contiguous voids. On the decoder side, we formulate recovery as prior-aware attribute inpainting. A Context-Aware Residual Interpolation (CARI) branch uses hash-grid prior predictions and neighboring residuals to build a robust baseline, while a lightweight two-layer graph neural network with cross-attention over hash-grid priors refines high-frequency attribute residuals. Attribute-wise confidence control falls back to interpolation when learned predictions are unreliable. Experiments under 20 percent random packet loss on BungeeNeRF, Mip-NeRF 360, and Tanks and Temples show that the proposed method substantially improves over no-concealment transmission and limits average PSNR degradation to about 3 dB relative to the lossless HAC++ reference.
