3D Gaussian Splatting (3DGS) has emerged as a powerful technique for novel view synthesis, demonstrating remarkable capability in high-fidelity scene reconstruction through its Gaussian primitive representations. However, the computational overhead induced by the massive number of primitives poses a significant bottleneck to training efficiency. To overcome this challenge, we propose Group Training, a simple yet effective strategy that organizes Gaussian primitives into manageable groups, optimizing training efficiency and improving rendering quality. This approach shows universal compatibility with existing 3DGS frameworks, including vanilla 3DGS and Mip-Splatting, consistently achieving accelerated training while maintaining superior synthesis quality. Extensive experiments reveal that our straightforward Group Training strategy achieves up to 30% faster convergence and improved rendering quality across diverse scenarios.
3D高斯投影（3D Gaussian Splatting, 3DGS）作为一种强大的新视角合成技术，通过高斯原语的表示展现了在高保真场景重建中的卓越能力。然而，由于高斯原语数量庞大，其计算开销显著，成为训练效率的主要瓶颈。 为解决这一问题，我们提出了一种简单而高效的策略——分组训练（Group Training），通过将高斯原语组织成可管理的分组来优化训练效率，并提升渲染质量。该方法具有与现有3DGS框架（包括基础3DGS和Mip-Splatting）的广泛兼容性，能够在加速训练的同时保持卓越的合成质量。 广泛的实验表明，我们的分组训练策略在多种场景下实现了高达30%的训练收敛加速，并在渲染质量上取得了进一步提升。这一方法不仅简便易用，还为3DGS技术在高效训练和优质合成之间提供了新的平衡点。