Best Practices for 2-Body Pose Forecasting (2023-04-12T00:00:00.000000Z)

TL;DR

This study confirms the positive impact of frequency input representations, space-time separable and fully-learnable interaction adjacencies for the encoding GCN and FC decoding and contributes a novel initialization procedure for the 2-body spatial interaction parameters of the encoder, which benefits performance and stability.

Abstract

The task of collaborative human pose forecasting stands for predicting the future poses of multiple interacting people, given those in previous frames. Predicting two people in interaction, instead of each separately, promises better performance, due to their body-body motion correlations. But the task has remained so far primarily unexplored.In this paper, we review the progress in human pose forecasting and provide an in-depth assessment of the single-person practices that perform best for 2-body collaborative motion forecasting. Our study confirms the positive impact of frequency input representations, space-time separable and fully-learnable interaction adjacencies for the encoding GCN and FC decoding. Other single-person practices do not transfer to 2-body, so the proposed best ones do not include hierarchical body modeling or attention-based interaction encoding.We further contribute a novel initialization procedure for the 2-body spatial interaction parameters of the encoder, which benefits performance and stability. Altogether, our proposed 2-body pose forecasting best practices yield a performance improvement of 21.9% over the state-of-the-art on the most recent ExPI dataset, whereby the novel initialization accounts for 3.5%. See our project page at https://www.pinlab.org/bestpractices2body

Authors

Alessandro Flaborea

3 papers

Alessio Sampieri

2 papers

Fabio Galasso

3 papers

TL;DR

Abstract

Authors

References59 items

Pose Forecasting in Industrial Human-Robot Collaboration

Back to MLP: A Simple Baseline for Human Motion Prediction

MotionMixer: MLP-based 3D Human Body Pose Forecasting

Progressively Generating Better Initial Guesses Towards Next Stages for High-Quality Human Motion Prediction

Social-Implicit: Rethinking Trajectory Prediction Evaluation and The Effectiveness of Implicit Maximum Likelihood Estimation

Multi-Person 3D Motion Prediction with Multi-Range Transformers

Space-Time-Separable Graph Convolutional Network for Pose Forecasting

MSR-GCN: Multi-Scale Residual Graph Convolution Networks for Human Motion Prediction

Multi-level Motion Attention for Human Motion Prediction

Multi-Person Extreme Motion Prediction

SGCN:Sparse Graph Convolution Network for Pedestrian Trajectory Prediction

AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent Forecasting

MagNet: A Neural Network for Directed Graphs

PI-Net: Pose Interacting Network for Multi-Person Monocular 3D Pose Estimation

History Repeats Itself: Human Motion Prediction via Motion Attention

Human Trajectory Forecasting in Crowds: A Deep Learning Perspective

Three-Dimensional Reconstruction of Human Interactions

DAG-Net: Double Attentive Graph Neural Network for Trajectory Forecasting

Directed Graph Convolutional Network

A Spatio-temporal Transformer for 3D Human Motion Prediction

EvolveGraph: Multi-Agent Trajectory Prediction with Dynamic Relational Reasoning

Dynamic Multiscale Graph Neural Networks for 3D Skeleton Based Human Motion Prediction

Symbiotic Graph Neural Networks for 3D Skeleton-Based Human Action Recognition and Motion Prediction

STGAT: Modeling Spatial-Temporal Interactions for Human Trajectory Prediction

Imitation Learning for Human Pose Prediction

Multi-Person 3D Human Pose Estimation from Monocular Images

Deep, Robust and Single Shot 3D Multi-Person Human Pose Estimation from Monocular Images

Learning Trajectory Dependencies for Human Motion Prediction

Action-Agnostic Human Pose Forecasting

Fast and Furious: Real Time End-to-End 3D Detection, Tracking and Motion Forecasting with a Single Convolutional Net

Convolutional Sequence to Sequence Model for Human Dynamics

An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling

Neural Relational Inference for Interacting Systems

Action Anticipation: Reading the Intentions of Humans and Robots

Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition

Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learning

Graph Attention Networks

PoseTrack: A Benchmark for Human Pose Estimation and Tracking

Convolutional Sequence to Sequence Learning

On Geometric Features for Skeleton-Based Action Recognition Using Multilayer LSTM Networks

Semi-Supervised Classification with Graph Convolutional Networks

All you need is a good init

Diffusion-Convolutional Neural Networks

Recurrent Network Models for Human Dynamics

Hierarchical recurrent neural network for skeleton based action recognition

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification

Adam: A Method for Stochastic Optimization

Human3.6M: Large Scale Datasets and Predictive Methods for 3D Human Sensing in Natural Environments

Anticipating human activities for reactive robotic response

ImageNet classification with deep convolutional neural networks

Understanding the difficulty of training deep feedforward neural networks

Nonrigid Structure from Motion in Trajectory Space

Human pose estimation from monocular images

Learning Progressive Joint Propagation for Human Motion Prediction

Graphs

CONVOLUTIONAL NEURAL NETWORKS

applicable license agreement with IEEE. Restrictions apply

Authorized licensed use limited to the terms of the applicable license agreement with IEEE

Field of Study

Journal Information

Name

Page