Outdoor Scene Extrapolation with Hierarchical Generative Cellular Automata
This work addresses the need for realistic, high-resolution 3D street environments for autonomous vehicle simulation, representing an incremental advance in scene extrapolation methods.
The paper tackles the problem of generating fine-grained 3D geometry from sparse LiDAR scans for autonomous vehicles, proposing hierarchical Generative Cellular Automata (hGCA) to extrapolate beyond scan limits, resulting in higher fidelity and completeness compared to state-of-the-art baselines in experiments on synthetic scenes and qualitative improvements on real-world data.
We aim to generate fine-grained 3D geometry from large-scale sparse LiDAR scans, abundantly captured by autonomous vehicles (AV). Contrary to prior work on AV scene completion, we aim to extrapolate fine geometry from unlabeled and beyond spatial limits of LiDAR scans, taking a step towards generating realistic, high-resolution simulation-ready 3D street environments. We propose hierarchical Generative Cellular Automata (hGCA), a spatially scalable conditional 3D generative model, which grows geometry recursively with local kernels following, in a coarse-to-fine manner, equipped with a light-weight planner to induce global consistency. Experiments on synthetic scenes show that hGCA generates plausible scene geometry with higher fidelity and completeness compared to state-of-the-art baselines. Our model generalizes strongly from sim-to-real, qualitatively outperforming baselines on the Waymo-open dataset. We also show anecdotal evidence of the ability to create novel objects from real-world geometric cues even when trained on limited synthetic content. More results and details can be found on https://research.nvidia.com/labs/toronto-ai/hGCA/.