Search papers, labs, and topics across Lattice.
2
4
4
2
Synthesizing new city driving environments from HD maps and visual cues allows autonomous vehicles to train for new locations without needing any labeled data.
Ditch the bounding boxes: DualDiff leverages Occupancy Ray-shape Sampling to generate driving scene videos with unprecedented fidelity, outperforming existing methods by a significant margin in both generation quality and downstream task performance.
