UnO: Unsupervised Occupancy Fields for Perception and Forecasting

Unsupervised 4D Occupancy (b) Rendered Point Cloud (c) BEV Semantic Occupancy Figure 1. We present UNO, a world model that learns to predict 3D occupancy (a) over time from unlabeled data. This model can be easily and effectively transferred to downstream tasks like point cloud forecasting (b), and bird's-eye view semantic occupancy (c).