[CVPR'24] GraphDreamer: a novel framework of generating compositional 3D scenes from scene graphs.
-
Updated
Jun 2, 2026 - Python
[CVPR'24] GraphDreamer: a novel framework of generating compositional 3D scenes from scene graphs.
[NeurIPS 2023] The repo of CommonScenes, a scene generation method powered by the diffusion model.
[RA-L 2026] Factorized Diffusion Policy
[ECCV 2026] An official implementation of Correlation-Weighted Multi-Reward Optimization for Compositional Generation
[ICASSP 2024] Language-free-Compositional-Action-Generation-via-Decoupling-Refinement
LoRA fine-tuning of OpenAI's GLIDE diffusion model on CelebA for compositional face generation. Combine and negate attributes, e.g. "a woman AND smiling AND NOT glasses", with composable diffusion guidance.
To associate your repository with the compositional-generation topic, visit your repo's landing page and select "manage topics."