AOGN-CZSL: An Attribute- and Object-Guided Network for Compositional Zero-Shot Learning

Jing Yang, Xingjiang Ma, Yuankai Wu, Chengjiang Li, Zhidong Su, Ji Xu, Yixiong Feng

Abstract: Highlights•Introducing dependency learning to improve zero-shot performance.•Introducing Attention Mechanisms to Handle Text Modality.•New graphic-textual multimodal learning space.•Performs well on widely used datasets.•Testing in the open world.