A pose-control module using self-supervised depth and forward plus inverse warping losses improves camera alignment in diffusion and autoregressive world models.
Monoocc: Digging into monocular semantic occupancy prediction,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
PosePilot: Steering Camera Pose for Generative World Models with Self-supervised Depth
A pose-control module using self-supervised depth and forward plus inverse warping losses improves camera alignment in diffusion and autoregressive world models.