Spatio-Temporal Graph Transformer Networks for Pedestrian Trajectory Prediction.
Cunjun YuXiao MaJiawei RenHaiyu ZhaoShuai YiPublished in: ECCV (12) (2020)
Keyphrases
- spatio temporal
- prediction accuracy
- trajectories of moving objects
- neural network ensemble
- graph structures
- graph structure
- object trajectories
- spatial and temporal
- complex networks
- edge weights
- graph representation
- average degree
- prediction model
- random walk
- small world
- structured data
- neural network
- moving objects databases
- moving object databases
- moving objects
- motion trajectories
- prediction error
- weighted graph
- fully connected
- connected components
- protein interaction networks
- degree distribution
- citation networks
- dynamic networks
- movement patterns
- overlapping communities
- directed acyclic graph
- network analysis
- betweenness centrality
- pedestrian detection
- fault diagnosis
- power system
- space time
- object detection
- fuzzy logic
- directed edges
- spatio temporal data
- power law
- graph model
- graph mining
- graph theory
- bipartite graph
- network structure
- computer vision
- social networks