# AiuniAI/Unique3D

[NeurIPS 2024] Unique3D: High-Quality and Efficient 3D Mesh Generation from a Single Image

Repository: https://github.com/AiuniAI/Unique3D
Canonical: https://ross.abutalabs.com/products/unique3d
Homepage: https://wukailu.github.io/Unique3D/
Language: Python
License: MIT
License Family: permissive
Topics: 3d-aigc, aigc, image-to-3d
Last push: 2025-07-17T21:29:05+00:00

## Health v2 (maintenance only)
Score: 38/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 32, release rhythm 35, longevity 58
- inputs: {"age_days": 825, "days_push": 412, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3579, forks 295 (observed 2026-08-28T04:08:10.894289+00:00)

## What it is
Unique3D is the official implementation of a NeurIPS 2024 paper that generates high-quality textured 3D meshes from a single image in about 30 seconds. It combines a multi-view diffusion model with normal maps and an ISOMER mesh reconstruction algorithm, runnable locally via Gradio, Docker, or ComfyUI.

## Use cases
- generate a 3d mesh from a single photo
- convert an image into a textured 3d model
- create 3d assets from single-view images for games
- reconstruct 3d geometry with normal maps from one image
- run image-to-3d generation locally on gpu
- integrate image-to-3d into comfyui workflows

## When to choose
- you need fast single-image to textured-mesh generation with high fidelity
- you want a local, self-hosted image-to-3d pipeline with GPU
- you need a research-grade NeurIPS-published image-to-3d model with weights available

## When to avoid
- you need training code or fine-tuning (not yet released)
- you lack a CUDA GPU or cannot handle heavy dependencies like mmcv and triton
- you need guaranteed production stability - the repo is still under construction

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing, graphics, simulation
- domain: artificial-intelligence, computer-vision, graphics, machine-learning, deep-learning
- platform: windows, python
- tags: image-to-3d, 3d-mesh-generation, multi-view-diffusion, textured-mesh, aigc, neurips-2024, gradio-demo, comfyui, linux, gpu, docker

## Member repositories
- AiuniAI/Unique3D (main) score 38

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:10.894289+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:34:00.062262+00:00, confidence not recorded.
  - readme: https://github.com/AiuniAI/Unique3D (fetched 2026-08-28T04:08:10.894289+00:00, sha 2b2be6d46753)
  - homepage: https://wukailu.github.io/Unique3D/ (fetched 2026-08-29T09:27:31.810622+00:00, sha ad797e79e331)
- Data as of 2026-08-30T08:39:29.467469+00:00.