Showing cs.CVShow all
2 papers · 1 filter
cs.CV2026
Foundation4D: Towards a unified perception and reconstruction model for autonomous vehicles
NVIDIA, :, Sérgio Agostinho +25
We introduce Foundation4D, a unified feed-forward model for geometry estimation, 3D detection, semantic occupancy, and Gaussian reconstruction from multi-camera driving video. A sh…
cs.CV2025
CuSfM: CUDA-Accelerated Structure-from-Motion
Jingrui Yu, Jun Liu, Kefei Ren +5
Efficient and accurate camera pose estimation forms the foundational requirement for dense reconstruction in autonomous navigation, robotic perception, and virtual simulation syste…