Deep relational self-Attention networks for scene graph generation.
Ping LiZhou YuYibing ZhanPublished in: Pattern Recognit. Lett. (2022)
Keyphrases
- video sequences
- relational databases
- small world
- d scene
- social networks
- single image
- dynamic networks
- community discovery
- input image
- directed acyclic graph
- three dimensional
- scene understanding
- graph model
- weighted graph
- graph theory
- average degree
- data model
- degree distribution
- random walk
- edge weights
- fully connected
- complex networks
- network analysis
- social graphs
- graph layout
- overlapping communities
- connected components
- real scenes
- graph representation
- graphical representation
- graph structure
- relational data
- bipartite graph
- directed edges
- structural patterns
- relational graph
- betweenness centrality
- network size
- graph structures
- relational structures
- path length
- graph databases
- complex scenes
- graph mining
- graph matching
- pattern mining
- network structure
- moving objects