junshutang/Make-It-3D

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1891 stars · 135 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Make-It-3D is a research codebase from an ICCV 2023 paper that creates high-fidelity 3D content from a single image using a 2D diffusion model prior. It optimizes a neural radiance field in a coarse stage and then refines it into textured point clouds for realistic 3D reconstruction.

Use cases

• generate a 3D model from a single photo
• reconstruct 3D geometry and textures from one image
• create 3D assets from text prompts
• edit textures of 3D objects with diffusion priors
• render 360-degree views of an object from one view
• research single-image 3D generation methods

When to choose

• you need research-grade single-image-to-3D generation with NeRF and diffusion priors
• you want to reproduce or build on an ICCV 2023 3D creation paper
• you have a GPU environment and want to experiment with text-to-3D or image-to-3D pipelines

When to avoid

• you need a production-ready or user-friendly 3D modeling tool
• you lack a CUDA-capable GPU or cannot set up heavy deep-learning dependencies
• you need a permissively licensed project - the repository has no license, so reuse is legally restricted

Facets

library · maturity maintenance

machine-learning deep-learning image-processing graphics computer-vision artificial-intelligence deep-learning graphics python 3d-generation diffusion-models nerf single-image-to-3d iccv-2023 research-code generative-ai gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="junshutang/Make-It-3D")