Dynamic Dense Graph Convolutional Network for Skeleton-based Human Motion Prediction

TL;DR

This paper introduces DD-GCN, a novel graph convolutional network that constructs dense, 4D adjacency graphs and employs dynamic message passing to improve skeleton-based human motion prediction accuracy.

Contribution

It presents a dynamic dense graph construction and message passing framework that adaptively models motion sequences, outperforming existing GCN methods.

Findings

Outperforms state-of-the-art GCN methods on benchmark datasets.

Effective in long-term and extremely long-term motion prediction.

Demonstrates the benefit of dynamic, data-driven message passing.

Abstract

Graph Convolutional Networks (GCN) which typically follows a neural message passing framework to model dependencies among skeletal joints has achieved high success in skeleton-based human motion prediction task. Nevertheless, how to construct a graph from a skeleton sequence and how to perform message passing on the graph are still open problems, which severely affect the performance of GCN. To solve both problems, this paper presents a Dynamic Dense Graph Convolutional Network (DD-GCN), which constructs a dense graph and implements an integrated dynamic message passing. More specifically, we construct a dense graph with 4D adjacency modeling as a comprehensive representation of motion sequence at different levels of abstraction. Based on the dense graph, we propose a dynamic message passing framework that learns dynamically from data to generate distinctive messages reflecting sample-specific relevance among nodes in the graph. Extensive experiments on benchmark Human 3.6M and CMU Mocap datasets verify the effectiveness of our DD-GCN which obviously outperforms state-of-the-art GCN-based methods, especially when using long-term and our proposed extremely long-term protocol.