A single transformer latent pretrained for 3D reconstruction, adapted with camera calibration and split attention, drives depth, detection, and occupancy for self-driving cameras.
Densebev: Transforming bev grid cells into 3d objects
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Geometry-Grounded Unified 3D Perception for Autonomous Driving
A single transformer latent pretrained for 3D reconstruction, adapted with camera calibration and split attention, drives depth, detection, and occupancy for self-driving cameras.