---
title: 3-D Vision & Neural Rendering
description: Point-cloud and mesh processing, differentiable 3D ops, NeRF and Gaussian-splatting tooling.
tags: [machine-learning, vision]
---

# 3-D Vision & Neural Rendering

Fetch when the target works with point clouds, meshes, or neural rendering. Each entry is one line and a documentation entry point; fetch build and API details from the entry point, never from memory. No entry is a recommendation.

## Tools

| Tool | One line | Docs |
|---|---|---|
| Open3D | 3D data processing: point clouds, meshes, visualization | <https://www.open3d.org/docs/> |
| PyTorch3D | differentiable 3D ops for PyTorch | <https://pytorch3d.org/> |
| NVIDIA Kaolin | 3D deep-learning research library | <https://kaolin.readthedocs.io/> |
| Nerfstudio | NeRF training and rendering pipelines | <https://docs.nerf.studio/> |
| gsplat | Gaussian-splatting rasterization library | <https://docs.gsplat.studio/> |
| tiny-cuda-nn | fast fused MLPs and hash encodings | <https://github.com/NVlabs/tiny-cuda-nn> |
| OpenPCDet | LiDAR 3D detection toolbox | <https://github.com/open-mmlab/OpenPCDet> |
| Trimesh | mesh loading and geometry processing | <https://trimesh.org/> |

## Gotchas

- PyVista is recorded with scientific visualization — see the [visualization-and-apps page](visualization-and-apps.md).