Interactive Humanoid: Online Full-Body Motion Reaction Synthesis with Social Affordance Canonicalization and Forecasting

TL;DR

This paper introduces an online full-body motion reaction synthesis method for humanoids, leveraging social affordance canonicalization and forecasting to generate realistic reactions in human-humanoid interactions, including object handling.

Contribution

It presents a novel task, datasets, and a unified approach that incorporates social affordance representation and forecasting for real-time humanoid reaction synthesis.

Findings

Effective generation of high-quality humanoid reactions.

Successful validation on multiple datasets including HHI, CoChair, Interhuman, and Chi3D.

Advancement over prior work by including object interaction and online reaction capability.

Abstract

We focus on the human-humanoid interaction task optionally with an object. We propose a new task named online full-body motion reaction synthesis, which generates humanoid reactions based on the human actor's motions. The previous work only focuses on human interaction without objects and generates body reactions without hand. Besides, they also do not consider the task as an online setting, which means the inability to observe information beyond the current moment in practical situations. To support this task, we construct two datasets named HHI and CoChair and propose a unified method. Specifically, we propose to construct a social affordance representation. We first select a social affordance carrier and use SE(3)-Equivariant Neural Networks to learn the local frame for the carrier, then we canonicalize the social affordance. Besides, we propose a social affordance forecasting scheme to enable the reactor to predict based on the imagined future. Experiments demonstrate that our approach can effectively generate high-quality reactions on HHI and CoChair. Furthermore, we also validate our method on existing human interaction datasets Interhuman and Chi3D.