AtlasLC: Fast Codec-Ready Compression of Object-Centric 3D Gaussian Splatting

2026-07-29

AI Summary

Existing 3D Gaussian splatting compression methods struggle to efficiently support the rapid packaging, transmission, and decoding of reusable foreground objects in extended reality (XR) applications, often overlooking deployment-critical factors such as depth and silhouette fidelity. This work presents the first deployment-aware compression framework tailored for object-level 3D Gaussian assets, operating directly on published assets without requiring source data or retraining. The approach integrates local competitive pruning, deterministic atlas packing, and lightweight single-pass sort-order-preserving transmission within a unified coordinate system to circumvent remapping bottlenecks. Experiments demonstrate up to a 25× acceleration in atlas preparation and a 5× end-to-end compression speedup. Compared to compact baselines, the method achieves 6–8% bitrate savings while maintaining comparable perceptual and geometric quality, ensuring low decoding latency and high rendering frame rates.

Abstract

3D Gaussian Splatting (3DGS) enables photorealistic novel-view synthesis with real-time rendering, but deploying compressed object-centric 3DGS in XR requires more than image-space rate-distortion. In practical XR asset pipelines, reusable objects are repeatedly packaged, transmitted, decoded, and instantiated, making asset-preparation cost, codec compatibility, decoding latency, and preservation of depth and silhouette cues first-class concerns. Existing 3DGS compression methods are largely developed for scene-scale captures and often rely on heavy layout generation or aggressive global pruning, assumptions that transfer poorly to semantically concentrated foreground objects. We present AtlasLC, a source-free, training-free compression pipeline for object-centric 3DGS that operates directly on released Gaussian assets, without original images, camera poses, or per-asset optimization. AtlasLC couples local-competition pruning with deterministic atlas packing to remove the mapping/remapping bottleneck while preserving object-wide foreground support; a lightweight single-pass sort-based conditional transport is used as a shared coordinate backbone for these stages. Across the evaluated assets, AtlasLC reduces atlas-preparation time by up to a factor of 25 and end-to-end compression time by up to a factor of 5, while offering a favorable deployment-aware balance of payload, decode latency, runtime FPS, and 3D geometry relative to the evaluated compressed baselines. Relative to similarly compact structured baselines, it uses about 6 to 8 percent fewer bits while maintaining comparable perceptual and geometric quality. These results show that object-centric 3DGS compression should be optimized for a deployment-aware operating point enabling scalable XR asset libraries.

Problem

• 3D Gaussian Splatting
• object-centric compression
• XR asset deployment
• codec compatibility
• depth preservation

Innovation

• object-centric 3DGS
• codec-ready compression
• local-competition pruning
• atlas packing
• deployment-aware optimization