发表机构
Sun Yat-sen University; Pengcheng Laboratory; Academy of Broadcasting Science, National Radio and Television Administration; Peking University Shenzhen Graduate School（中山大学; 鹏城实验室; 国家广播电视总局广播电视科学研究院; 北京大学深圳研究生院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出COSA-GS，通过锚点级因果分解构建无空间聚合的上下文模型，结合量化感知训练和整数推理，实现跨平台比特级一致的3DGS高效压缩。
AI 中文摘要
3D高斯泼溅（3DGS）能够实现高质量的新视角合成，但需要大量的存储空间。现有的压缩方法通常依赖于对不规则3D表示进行空间上下文建模，这增加了训练和编码的复杂性。同时，浮点上下文推理可能在不同平台上引入数值不一致性，导致熵解码失败。为解决这些实际挑战，我们提出了COSA-GS，通过锚点级因果分解构建上下文，无需空间聚合。具体而言，我们利用每个锚点坐标导出的几何上下文来建模一个紧凑的可学习锚点潜变量。随后，将锚点潜变量与几何上下文融合，形成用于属性编码的锚点上下文。所得到的上下文模型架构简单，仅由线性变换和激活函数组成。我们使用率-失真优化结合自适应高斯剪枝来训练COSA-GS。此外，我们为上下文模型开发了量化感知训练和整数推理，以实现熵解码符号在跨平台上的比特级一致性。实验表明，COSA-GS在保持快速且一致的跨平台解码的同时，实现了最先进的压缩性能，为实用的3DGS压缩提供了一个简单而有效的框架。代码可在该https URL获取。
英文摘要
3D Gaussian Splatting (3DGS) enables high-quality novel-view synthesis but requires substantial storage. Existing compression methods often rely on spatial context modeling over irregular 3D representations, increasing the complexity of training and coding. Meanwhile, floating-point context inference can introduce numerical inconsistencies across platforms, causing entropy-decoding failures. To address these practical challenges, we propose COSA-GS, which constructs context without spatial aggregation through anchor-wise causal factorization. Specifically, we use geometry context derived from each anchor's coordinates to model a compact learnable anchor latent. The anchor latent is then fused with the geometry context to form an anchor context for attribute coding. The resulting context model features a simple architecture composed solely of linear transformations and activations. We train COSA-GS using rate--distortion optimization with adaptive Gaussian pruning. Further, we develop quantization-aware training and integer inference for the context model to achieve bit-exact consistency of entropy-decoded symbols across platforms. Experiments demonstrate that COSA-GS achieves state-of-the-art compression performance while retaining fast and consistent cross-platform decoding, providing a simple yet effective framework for practical 3DGS compression. Code is available at https://github.com/pengpeng-yu/COSA-GS.
