Faster and Better 3D Splatting via Group Training
This incremental improvement addresses training efficiency for 3D scene reconstruction in novel view synthesis.
The paper tackles the computational bottleneck in 3D Gaussian Splatting by proposing Group Training, which organizes primitives into groups to optimize efficiency, achieving up to 30% faster convergence and improved rendering quality.
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.