#3D Generation

Normal-Depth Diffusion Model provides an adaptable method for creating detailed 3D models from text prompts. It includes extensive inference and training codes, pre-trained models for normal, multi-view, and albedo diffusion, and offers a multi-view dataset from Objaverse, facilitating detailed 3D generation. The project continuously evolves with new updates, offering accessible training datasets and integration with leading AI frameworks to enhance accuracy in text-to-3D rendering.

GeoDream enhances precision in 3D generation by integrating 3D and 2D priors, ensuring realistic, high-resolution outputs and consistent geometric structures. The Uni3D-score metric advances semantic coherence evaluation in 3D. Recent integrations with ThreeStudio and Stable-Zero123 extend its capabilities. Designed for research efficiency, GeoDream supports dual virtual environments, optimizing performance on NVIDIA GPUs with reduced VRAM need, and provides robust texture and mesh export options.

Discover techniques for converting text and images into 3D models using advanced 2D priors such as stable diffusion and CLIP. This repository showcases various methods including zero-shot text-guided generation and multi-view consistent image transformation, presenting developments since 2022. As research progresses, directly training on 3D data models offers promising results, with comprehensive listings of the latest advancements in text and image-guided 3D creation.

RichDreamer applies a Normal-Depth Diffusion Model to convert text into detailed 3D images, suitable for various applications. The model, accepted at CVPR2024, supports MultiView-ND and Albedo Diffusion for enhanced detail. It includes installation instructions and pre-trained weights, and works efficiently with various GPUs. Resources are available on ModelScope's 3D Object Generation platform.

Zero123++ v1.2 enhances multi-view image synthesis from a single input, emphasizing 3D generation and improved handling of camera settings. With the addition of a ControlNet normal generator, it achieves better mask accuracy. The model is easy to use with tools such as torch and diffusers, making it efficient for VRAM. Available under Apache 2.0 and CC-BY-NC 4.0 licenses, it can be accessed for non-commercial purposes on Hugging Face. Discover multi-view synthesis with included scripts and demos.

This project offers a detailed curation of Text-to-3D and Diffusion-to-3D methods inspired by awesome-NeRF, featuring numerous research papers on converting text to 3D models using techniques like Neural Radiance Fields (NeRF) and advanced diffusion models. It is regularly updated with new resources including project page links, video tutorials, and cited literature with code repositories, providing a valuable resource for exploring innovative 3D synthesis developments. Applications range from object generation to novel view synthesis, serving as an informative guide for researchers and enthusiasts.