Nathan Piasco

HELIOS outperforms traditional relighting techniques by enabling realistic scene illumination transitions without paired training data, transforming how we approach outdoor lighting challenges.

Cyclone achieves realistic weather editing without paired data, revolutionizing how we train perception models for autonomous driving.

DrivingVoxels achieves faster and more efficient dynamic scene reconstruction by leveraging independent octrees for rigid objects and a static background, outperforming existing methods in both speed and accuracy.

Generate diverse, photorealistic, and large-scale driving scenes without per-scene optimization using a novel semantic voxel-guided diffusion approach.