Visual-Semantic Graph Attention Networks for Human-Object Interaction Detection
Visual-Semantic Graph Attention Networks for Human-Object Interaction Detection
In scene understanding, robots benefit from not only detecting individual scene instances but also from learning their possible interactions. Human-Object Interaction (HOI) Detection infers the action predicate on a <human, predicate, object> triplet. Contextual information has been found critical in inferring interactions. However, most works only use local features from …