# lizhihao6/Sparc3D

Official repo for paper "Sparse Representation and Construction for High-Resolution 3D Shapes Modeling".

Repository: https://github.com/lizhihao6/Sparc3D
Canonical: https://ross.abutalabs.com/products/sparc3d
Homepage: https://lizhihao6.github.io/Sparc3D/
License Family: other
Last push: 2025-06-16T05:29:19+00:00

## Health v2 (maintenance only)
Score: 31/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 27, release rhythm 35, longevity 33
- inputs: {"age_days": 469, "days_push": 443, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1353, forks 72 (observed 2026-08-28T04:04:28.677228+00:00)

## What it is
Sparc3D is the official implementation of a research framework for high-resolution 3D shape modeling, combining a sparse deformable marching cubes representation (Sparcubes) with a sparse-convolution VAE (Sparconv-VAE). It enables near-lossless 1024³ 3D reconstruction and integrates with latent diffusion models for image-to-3D generation.

## Use cases
- generate 3D models from a single image
- reconstruct high-resolution meshes from raw geometry
- build a VAE for 3D shape latent diffusion
- convert meshes to sparse volumetric representations
- research on sparse convolutional networks for 3D
- preserve fine details in 3D asset generation

## When to choose
- you need state-of-the-art high-fidelity 3D reconstruction or image-to-3D generation
- you are researching sparse convolution VAEs or latent diffusion for 3D shapes
- you need to handle open surfaces, disconnected components, or intricate geometry

## When to avoid
- you need a production-ready tool with a stable API and license
- you want simple, low-poly 3D modeling without GPU resources
- code release is still pending approval, so you need something immediately runnable

## Facets
- artifact type: library
- maturity: experimental
- function: machine-learning, deep-learning, image-processing, graphics, simulation
- domain: artificial-intelligence, deep-learning, computer-vision, graphics, machine-learning
- platform: python, cross-platform
- tags: 3d-generation, 3d-reconstruction, sparse-convolution, variational-autoencoder, latent-diffusion, mesh-processing, research-code, image-to-3d, gpu, linux

## Member repositories
- lizhihao6/Sparc3D (main) score 31

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:28.677228+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 2026-08-30T04:42:06.726524+00:00, confidence not recorded.
  - readme: https://github.com/lizhihao6/Sparc3D (fetched 2026-08-28T04:04:28.677228+00:00, sha cb62f80a7a03)
  - homepage: https://lizhihao6.github.io/Sparc3D/ (fetched 2026-08-29T12:00:56.974033+00:00, sha 6fef7888f54e)
- Data as of 2026-08-30T08:39:29.467469+00:00.