# nywang16/Pixel2Mesh

Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images. In ECCV2018.

Repository: https://github.com/nywang16/Pixel2Mesh
Canonical: https://ross.abutalabs.com/products/pixel2mesh
Homepage: http://openaccess.thecvf.com/content_ECCV_2018/papers/Nanyang_Wang_Pixel2Mesh_Generating_3D_ECCV_2018_paper.pdf
Language: Python
License: Apache-2.0
License Family: permissive
Topics: 3d, mesh, reconstruction
Last push: 2021-12-15T13:35:18+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2959, "days_push": 1722, "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 1765, forks 305 (observed 2026-08-28T04:05:33.047777+00:00)

## What it is
Pixel2Mesh is the official TensorFlow implementation of the ECCV 2018 paper that generates 3D mesh models from a single RGB image using a graph convolutional encoder-decoder. It includes pre-trained models, a demo script, and CUDA implementations of Chamfer Distance and Earth Mover's Distance for training and evaluation.

## Use cases
- reconstruct a 3d mesh from a single photo
- generate 3d models from rgb images
- run pixel2mesh demo on a picture
- train a single-image 3d reconstruction model
- evaluate 3d shape reconstruction with chamfer distance and emd
- convert an image of an object into an obj mesh file

## When to choose
- you need to reproduce or build on the ECCV 2018 Pixel2Mesh paper
- you want a pre-trained model to turn single images into 3D meshes
- you need CUDA implementations of CD and EMD loss metrics for point cloud research

## When to avoid
- you need a maintained production-ready library - the code targets Python 2.7 and TensorFlow 1.x
- you want multi-view or video-based 3D reconstruction
- you lack a CUDA-enabled GPU

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, graphics, simulation
- domain: computer-vision, machine-learning, deep-learning, graphics, artificial-intelligence
- platform: python
- tags: 3d-reconstruction, mesh-generation, single-image, tensorflow, eccv2018, shapenet, research-code, linux, gpu

## Member repositories
- nywang16/Pixel2Mesh (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:33.047777+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 2026-08-30T03:26:37.119301+00:00, confidence not recorded.
  - readme: https://github.com/nywang16/Pixel2Mesh (fetched 2026-08-28T04:05:33.047777+00:00, sha 17870fe1db22)
- Data as of 2026-08-30T08:39:29.467469+00:00.