HOLODECK 2.0 generates and edits stylistically varied 3D scenes from text by combining VLM-based scene analysis and layout planning with generative 3D assets, outperforming the retrieval-based HOLODECK baseline in human and CLIP evaluations.
Title resolution pending
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
-
HOLODECK 2.0: Vision-Language-Guided 3D World Generation with Editing
HOLODECK 2.0 generates and edits stylistically varied 3D scenes from text by combining VLM-based scene analysis and layout planning with generative 3D assets, outperforming the retrieval-based HOLODECK baseline in human and CLIP evaluations.