EvoSTGAT: Evolving spatiotemporal graph attention networks for pedestrian trajectory prediction.
Haowen TangPing WeiJiapeng LiNanning ZhengPublished in: Neurocomputing (2022)
Keyphrases
- prediction accuracy
- average degree
- evolving graphs
- random walk
- neural network ensemble
- graph model
- network analysis
- weighted graph
- prediction model
- edge weights
- prediction algorithm
- graph structures
- highly connected
- small world
- graph representation
- network size
- social networks
- object detection
- directed edges
- graph theory
- spatio temporal
- betweenness centrality
- fully connected
- spatial and temporal
- graph structure
- space time
- overlapping communities
- community discovery
- neural network
- continuously moving
- graph layout
- social graphs
- location prediction
- complex networks
- connected components
- directed graph
- heterogeneous networks
- visual attention
- graph theoretic
- graph partitioning
- graph databases