GaussianPrediction: Dynamic 3D Gaussian Prediction for Motion Extrapolation and Free View Synthesis

TL;DR

GaussianPrediction introduces a novel 3D Gaussian framework for dynamic scene modeling, enabling accurate future scenario prediction and free view synthesis from arbitrary viewpoints using video observations.

Contribution

It presents a new 3D Gaussian-based method with deformation modeling, motion distillation, and GCNs for efficient and realistic future environment prediction.

Findings

Outperforms existing methods on synthetic datasets

Achieves high-quality free view synthesis in real-world scenes

Effectively models irreversible deformations in dynamic environments

Abstract

Forecasting future scenarios in dynamic environments is essential for intelligent decision-making and navigation, a challenge yet to be fully realized in computer vision and robotics. Traditional approaches like video prediction and novel-view synthesis either lack the ability to forecast from arbitrary viewpoints or to predict temporal dynamics. In this paper, we introduce GaussianPrediction, a novel framework that empowers 3D Gaussian representations with dynamic scene modeling and future scenario synthesis in dynamic environments. GaussianPrediction can forecast future states from any viewpoint, using video observations of dynamic scenes. To this end, we first propose a 3D Gaussian canonical space with deformation modeling to capture the appearance and geometry of dynamic scenes, and integrate the lifecycle property into Gaussians for irreversible deformations. To make the prediction feasible and efficient, a concentric motion distillation approach is developed by distilling the scene motion with key points. Finally, a Graph Convolutional Network is employed to predict the motions of key points, enabling the rendering of photorealistic images of future scenarios. Our framework shows outstanding performance on both synthetic and real-world datasets, demonstrating its efficacy in predicting and rendering future environments.