AeroScene uses hierarchy-aware tokenization and multi-branch feature extraction in a diffusion model to synthesize physically plausible 3D scenes for aerial robotics, outperforming prior methods and enabling a public dataset of over 1,000 scenes for simulators like NVIDIA Isaac Sim.
Sdedit: Guided image synthesis and editing with stochastic differential equations,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
AeroScene: Progressive Scene Synthesis for Aerial Robotics
AeroScene uses hierarchy-aware tokenization and multi-branch feature extraction in a diffusion model to synthesize physically plausible 3D scenes for aerial robotics, outperforming prior methods and enabling a public dataset of over 1,000 scenes for simulators like NVIDIA Isaac Sim.