GenieDrive: Towards Physics-Aware Driving World Model with 4D Occupancy Guided Video Generation

TL;DR

GenieDrive introduces a physics-informed framework for driving video generation that leverages 4D occupancy, a specialized VAE, and attention mechanisms to produce controllable, multi-view, and physically consistent driving videos efficiently.

Contribution

The paper presents a novel physics-aware driving video generation framework using 4D occupancy, a compact VAE, and attention modules, achieving improved accuracy and efficiency over prior methods.

Findings

7.2% improvement in forecasting mIoU

20.7% reduction in FVD for video quality

41 FPS inference speed with 3.47 M parameters

Abstract

Physics-aware driving world model is essential for drive planning, out-of-distribution data synthesis, and closed-loop evaluation. However, existing methods often rely on a single diffusion model to directly map driving actions to videos, which makes learning difficult and leads to physically inconsistent outputs. To overcome these challenges, we propose GenieDrive, a novel framework designed for physics-aware driving video generation. Our approach starts by generating 4D occupancy, which serves as a physics-informed foundation for subsequent video generation. 4D occupancy contains rich physical information, including high-resolution 3D structures and dynamics. To facilitate effective compression of such high-resolution occupancy, we propose a VAE that encodes occupancy into a latent tri-plane representation, reducing the latent size to only 58% of that used in previous methods. We further introduce Mutual Control Attention (MCA) to accurately model the influence of control on occupancy evolution, and we jointly train the VAE and the subsequent prediction module in an end-to-end manner to maximize forecasting accuracy. Together, these designs yield a 7.2% improvement in forecasting mIoU at an inference speed of 41 FPS, while using only 3.47 M parameters. Additionally, a Normalized Multi-View Attention is introduced in the video generation model to generate multi-view driving videos with guidance from our 4D occupancy, significantly improving video quality with a 20.7% reduction in FVD. Experiments demonstrate that GenieDrive enables highly controllable, multi-view consistent, and physics-aware driving video generation.