ARMFlow: AutoRegressive MeanFlow for Online 3D Human Reaction Generation

TL;DR

ARMFlow is a novel autoregressive framework for online 3D human reaction generation that achieves high fidelity, real-time inference, and adaptability by modeling temporal dependencies and reducing error accumulation.

Contribution

It introduces ARMFlow with Bootstrap Contextual Encoding and a causal encoder, enabling online generation with improved accuracy and low latency, surpassing existing methods.

Findings

30% improvement in FID over existing online methods

State-of-the-art offline performance with ReMFlow

Real-time, high-fidelity online generation

Abstract

3D human reaction generation faces three main challenges:(1) high motion fidelity, (2) real-time inference, and (3) autoregressive adaptability for online scenarios. Existing methods fail to meet all three simultaneously. We propose ARMFlow, a MeanFlow-based autoregressive framework that models temporal dependencies between actor and reactor motions. It consists of a causal context encoder and an MLP-based velocity predictor. We introduce Bootstrap Contextual Encoding (BSCE) in training, encoding generated history instead of the ground-truth ones, to alleviate error accumulation in autoregressive generation. We further introduce the offline variant ReMFlow, achieving state-of-the-art performance with the fastest inference among offline methods. Our ARMFlow addresses key limitations of online settings by: (1) enhancing semantic alignment via a global contextual encoder; (2) achieving high accuracy and low latency in a single-step inference; and (3) reducing accumulated errors through BSCE. Our single-step online generation surpasses existing online methods on InterHuman and InterX by about 30% in FID, while matching offline state-of-the-art performance despite using only partial sequence conditions.