# hiroharu-kato/neural_renderer

"Neural 3D Mesh Renderer" (CVPR 2018) by H. Kato, Y. Ushiku, and T. Harada.

Repository: https://github.com/hiroharu-kato/neural_renderer
Canonical: https://ross.abutalabs.com/products/hiroharu-kato-neural_renderer
Homepage: http://hiroharu-kato.com/projects_en/neural_renderer.html
Language: Python
License: MIT
License Family: permissive
Last push: 2022-03-21T13:31:40+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3206, "days_push": 1626, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1164, forks 157 (observed 2026-08-28T04:03:49.874415+00:00)

## What it is
A differentiable 3D mesh renderer implemented in Chainer, accompanying the CVPR 2018 paper 'Neural 3D Mesh Renderer'. It enables gradient-based optimization over 3D mesh vertices, textures, and camera parameters through the rendering process.

## Use cases
- render 3d meshes with gradients for deep learning
- optimize 3d mesh vertices from images
- optimize textures via differentiable rendering
- estimate camera pose by gradient descent
- single-image 3d mesh reconstruction research

## When to choose
- doing research on differentiable rendering or 3D reconstruction from images
- you need gradients flowing through a rasterization pipeline in Chainer
- reproducing CVPR 2018 neural rendering baselines

## When to avoid
- you work in PyTorch - use the community re-implementations instead
- you need a production-grade or actively maintained renderer
- you need full-featured rasterization like ray tracing or PBR shading

## Facets
- artifact type: library
- maturity: maintenance
- function: graphics, machine-learning, simulation
- domain: computer-vision, graphics, deep-learning
- platform: python
- tags: differentiable-rendering, 3d-mesh, chainer, cvpr-2018, computer-graphics, gpu, linux

## Member repositories
- hiroharu-kato/neural_renderer (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:49.874415+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 2026-08-30T06:29:40.652872+00:00, confidence not recorded.
  - readme: https://github.com/hiroharu-kato/neural_renderer (fetched 2026-08-28T04:03:49.874415+00:00, sha e59e5bfbd3ad)
- Data as of 2026-08-30T08:39:29.467469+00:00.