# PGformer: Proxy-Bridged Game Transformer for Multi-Person Highly Interactive Extreme Motion Prediction

arXiv: https://arxiv.org/abs/2306.03374 · HTML: https://mlsanity.com/paper/2306.03374/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 27 references that are in the map.
- First posted on arXiv 6 Jun 2023.

## Builds on (most cited first, 20 of 20)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [On human motion prediction using recurrent neural networks](https://mlsanity.com/paper/1705.02445/) (2017)
- [Social-STGCNN: A Social Spatio-Temporal Graph Convolutional Neural Network for Human Trajectory Prediction](https://mlsanity.com/paper/2002.11927/) (2020)
- [Learning Trajectory Dependencies for Human Motion Prediction](https://mlsanity.com/paper/1908.05436/) (2019)
- [Single-Shot Multi-Person 3D Pose Estimation From Monocular RGB](https://mlsanity.com/paper/1712.03453/) (2017)
- [History Repeats Itself: Human Motion Prediction via Motion Attention](https://mlsanity.com/paper/2007.11755/) (2020)
- [A Spatio-temporal Transformer for 3D Human Motion Prediction](https://mlsanity.com/paper/2004.08692/) (2020)
- [MSR-GCN: Multi-Scale Residual Graph Convolution Networks for Human Motion Prediction](https://mlsanity.com/paper/2108.07152/) (2021)
- [Structured Prediction Helps 3D Human Motion Modelling](https://mlsanity.com/paper/1910.09070/) (2019)
- [Uncertainty-aware Short-term Motion Prediction of Traffic Actors for Autonomous Driving](https://mlsanity.com/paper/1808.05819/) (2018)
- [Space-Time-Separable Graph Convolutional Network for Pose Forecasting](https://mlsanity.com/paper/2110.04573/) (2021)
- [Multi-Person Extreme Motion Prediction](https://mlsanity.com/paper/2105.08825/) (2021)
- [Multi-Person 3D Motion Prediction with Multi-Range Transformers](https://mlsanity.com/paper/2111.12073/) (2021)
- [TRiPOD: Human Trajectory and Pose Dynamics Forecasting in the Wild](https://mlsanity.com/paper/2104.04029/) (2021)
- [Spatio-Temporal Gating-Adjacency GCN for Human Motion Prediction](https://mlsanity.com/paper/2203.01474/) (2022)
- [Pose Transformers (POTR): Human Motion Prediction with Non-Autoregressive Transformers](https://mlsanity.com/paper/2109.07531/) (2021)
- [Skeleton-Parted Graph Scattering Networks for 3D Human Motion Prediction](https://mlsanity.com/paper/2208.00368/) (2022)
- [Trajectory-Aware Body Interaction Transformer for Multi-Person Pose Forecasting](https://mlsanity.com/paper/2303.05095/) (2023)
- [SoMoFormer: Multi-Person Pose Forecasting with Transformers](https://mlsanity.com/paper/2208.14023/) (2022)
- [Best Practices for 2-Body Pose Forecasting](https://mlsanity.com/paper/2304.05758/) (2023)

## Built on by (most cited first, 1 of 1)

- [Ponimator: Unfolding Interactive Pose for Versatile Human-human Interaction Animation](https://mlsanity.com/paper/2510.14976/) (2025)

## Abstract (arXiv)

Multi-person motion prediction is a challenging task, especially for real-world scenarios of highly interacted persons. Most previous works have been devoted to studying the case of weak interactions (e.g., walking together), in which typically forecasting each human pose in isolation can still achieve good performances. This paper focuses on collaborative motion prediction for multiple persons with extreme motions and attempts to explore the relationships between the highly interactive persons' pose trajectories. Specifically, a novel cross-query attention (XQA) module is proposed to bilaterally learn the cross-dependencies between the two pose sequences tailored for this situation. A proxy unit is additionally introduced to bridge the involved persons, which cooperates with our proposed XQA module and subtly controls the bidirectional spatial information flows. These designs are then integrated into a Transformer-based architecture and the resulting model is called Proxy-bridged Game Transformer (PGformer) for multi-person interactive motion prediction. Its effectiveness has been evaluated on the challenging ExPI dataset, which involves highly interactive actions. Our PGformer consistently outperforms the state-of-the-art methods in both short- and long-term predictions by a large margin. Besides, our approach can also be compatible with the weakly interacted CMU-Mocap and MuPoTS-3D datasets and extended to the case of more than 2 individuals with encouraging results.