OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Interdiff

Interdiff is an official PyTorch research implementation of a diffusion-based method for generating and anticipating 3D human-object interactions, with physics-informed correction of predicted motion.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Interdiff — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Interdiff when a research project needs a PyTorch implementation focused on anticipating 3D human-object interactions, including whole-body movement and dynamic objects.

Who it is for

  • Researchers studying 3D human-object interaction and motion prediction.
  • Animation researchers prototyping interactions between whole-body human motion and dynamic objects.

Documented uses

  • Explore predictions of future human and object motion together.
  • Prototype research workflows for 3D interactions involving moving objects.
  • Use provided training and evaluation code and checkpoints as a basis for research.
Before you choose

Requirements & limits

The repository is public and its code is distributed under the MIT License. Dependencies and datasets may have separate licenses.

  • The repository provides training and evaluation code, checkpoints, and data-preparation materials.
  • Setup documentation offers Python 3.7 and Python 3.8 dependency routes; some packages may require source builds.
  • The project depends on libraries and datasets with their own licensing terms.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare an Interdiff Python environment

Create an environment using one of the repository's documented dependency specifications.

Before you start

  • Python 3.7 for the requirements.txt-based setup, or Python 3.8 for the environment.yml alternative.
  • Some packages, including psbody-mesh and human-body-prior, may need to be built from source.

Steps

  1. Choose the requirements.txt-based Python 3.7 setup or the Python 3.8 environment.yml setup.

  2. For the latter, use the documented command: conda env create -f environment.yml.

  3. Consult the detailed implementation README for additional implementation instructions.

Expected outcome

A project environment prepared from one of the repository's stated dependency specifications.

Guide limitations

  • The documentation notes that some packages may need source builds.
  • This setup guidance does not cover running training or evaluation.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
Sirui-Xu/InterDiff · Official homepage title · retained in its source language
How we review sourcesSuggest a correction