Implicit 4D Gaussian Splatting for Fast Motion with Large Inter-Frame Displacements

2026-07-14

AI Summary

This work addresses the challenge that existing 4D Gaussian splatting methods struggle to accurately model Gaussian attributes under large inter-frame displacements caused by fast motion, often leading to the loss of moving objects. To overcome this limitation, the authors propose SPIN-4DGS, which introduces a spatiotemporal positional implicit representation mechanism. Instead of explicitly modeling temporal displacements or performing full spatiotemporal optimization, SPIN-4DGS employs a lightweight feedforward network to directly predict Gaussian attributes from explicitly sampled spatiotemporal coordinates, enabling shared representations across Gaussians. This approach significantly enhances reconstruction stability and quality under rapid motion, demonstrating superior performance on CMU Panoptic sports scenes—e.g., achieving a 1.83 dB PSNR gain and a corresponding improvement in SSIM in the Basketball sequence.

Abstract

Recent 4D Gaussian Splatting (4DGS) methods often fail under fast motion with large inter-frame displacements, where Gaussian attributes are poorly learned during training, and fast-moving objects are often lost from the reconstruction. In this work, we introduce Spatiotemporal Position Implicit Network for 4DGS, coined SPIN-4DGS, which learns Gaussian attributes from explicitly collected spatiotemporal positions rather than modeling temporal displacements, thereby enabling more faithful splatting under fast motions with large inter-frame displacements. To avoid the heavy memory overhead of explicitly optimizing attributes across all spatiotemporal positions, we instead predict them with a lightweight feed-forward network trained under a rasterization-based reconstruction loss. Consequently, SPIN-4DGS learns shared representations across Gaussians, effectively capturing spatiotemporal consistency and enabling stable high-quality Gaussian splatting even under challenging motions. Across extensive experiments, SPIN-4DGS consistently achieves higher fidelity under large displacements, with clear improvements in PSNR and SSIM on challenging sports scenes from the CMU Panoptic dataset. For example, SPIN-4DGS notably outperforms the strongest baseline, D3DGS, by achieving +1.83 higher PSNR on the Basketball scene.

Problem

• 4D Gaussian Splatting
• fast motion
• large inter-frame displacements
• reconstruction failure
• temporal consistency

Innovation

• 4D Gaussian Splatting
• implicit representation
• spatiotemporal consistency
• large inter-frame displacement
• fast motion reconstruction