EEmbodied AI Hub
ProjectFeaturedIntermediate

3D Diffusion Policy (DP3)

Point-cloud / 3D-aware diffusion policies for robot manipulation — popular open follow-up to DP.

Why: Go-to open 3D diffusion policy codebase

Official / sourcePaperBack to Project
Imitation learning扩散策略3DOpen source

Overview

3D Diffusion Policy (DP3) extends Diffusion Policy with point-cloud / 3D-aware conditioning for manipulation. It is one of the most cloned open follow-ups when RGB-only DP is not enough for geometry-heavy tasks.

Expect heavier preprocessing (point clouds, cameras) than 2D DP. Start from official configs and a single task before multi-task training.

Learning Path: advanced imitation option after classic Diffusion Policy.

Who it is for

Researchers, students, and builders comparing open embodied stacks.

Key highlights

  • Point-cloud / 3D-aware diffusion policies for robot manipulation — popular open follow-up to DP.
  • Type: project; categories: framework
  • Marked recommended on the hub

When to use

  • You are mapping the open embodied AI landscape
  • You need a starting URL and judgment note before deep-diving

When not to use

  • You need production SLAs or certified industrial support only
  • License or hardware constraints are incompatible with your deployment

Getting started

  1. 1Open the official URL and read README install pins.
  2. 2Skim license and citation requirements.
  3. 3Run the smallest official example before scaling experiments.
  4. 4Cross-link related hub resources (sim / data / policy) for a full stack.

Install / setup

git clone https://github.com/YanjieZe/3D-Diffusion-Policy.git

Papers & reading

How it compares

Caveats & pitfalls

  • Hub cards are curated summaries — always verify upstream docs.
  • Stars and dates drift; treat extra.stars as approximate.

Content reviewed 2026-07-28