Retrieval-Augmented Score Distillation for Text-to-3D Generation

Authors: Junyoung Seo, Susung Hong, Wooseok Jang, Inès Hyeonsu Kim, Min-Seop Kwak, Doyup Lee, Seungryong Kim

ICML 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct extensive experiments to demonstrate that Re Dream exhibits superior quality with increased geometric consistency.
Researcher Affiliation Collaboration 1Korea Univeristy, Seoul, Korea 2Runway, New York, USA. Correspondence to: Seungryong Kim <seungryong kim@korea.ac.kr>, Doyup Lee <doyup@runwayml.com>.
Pseudocode No The paper describes its methods using mathematical formulations and textual explanations, but does not provide structured pseudocode or algorithm blocks.
Open Source Code Yes Project page is available at https: //ku-cvlab.github.io/Re Dream/.
Open Datasets Yes We utilize 3D assets from Objaverse 1.0 (Deitke et al., 2023b) dataset and corresponding captions with the help of Cap3D (Luo et al., 2023).
Dataset Splits No The paper focuses on optimizing 3D representations rather than traditional model training with dataset splits. No specific training, validation, or test splits are provided for model training.
Hardware Specification Yes Our experiments were conducted on an NVIDIA RTX A6000 GPU, with a total of 20,000 iterations of optimization for generation.
Software Dependencies Yes For all our experiments, Instant-NGP (M uller et al., 2022) is used for our Ne RF backbone and Stable Diffusion v2 (Rombach et al., 2022b) as the 2D prior.
Experiment Setup Yes Our experiments were conducted on an NVIDIA RTX A6000 GPU, with a total of 20,000 iterations of optimization for generation. For all our experiments, Instant-NGP (M uller et al., 2022) is used for our Ne RF backbone and Stable Diffusion v2 (Rombach et al., 2022b) as the 2D prior. For our method, we retrieve 3 assets and render our retrieved data with 100 uniformly sampled camera poses.