Skeleton-Parted Graph Scattering Networks for 3D Human Motion Prediction.
Maosen LiSiheng ChenZijing ZhangLingxi XieQi TianYa ZhangPublished in: CoRR (2022)
Keyphrases
- prediction accuracy
- average degree
- protein function prediction
- community discovery
- prediction algorithm
- prediction model
- edge weights
- graph matching algorithm
- fully connected
- graph structures
- neural network ensemble
- random walk
- overlapping communities
- dynamic networks
- small world
- prediction error
- structured data
- social networks
- graph structure
- graph model
- graph representation
- shape representation
- directed graph
- human motion tracking
- highly connected
- directed acyclic graph
- network size
- community structure
- random graphs
- path length
- heterogeneous networks
- weighted graph
- graph mining
- betweenness centrality
- graph layout
- link formation
- network properties
- clustering coefficient
- topological features
- power law
- graph theoretic
- graph databases
- bipartite graph
- shape analysis
- connected components
- complex networks
- network structure