Key points are not available for this paper at this time.
Compositional Zero-Shot Learning (CZSL) aims to recognize novel concepts from known sub-concepts. However, it is still challenging since the intricate interaction between sub-concepts is entangled with their corresponding visual features, which affects the recognition accuracy of concepts. Besides, the domain gap between training and testing data leads to the model poor generalization. In this paper, we tackle these problems by exploring Partial Semantic Consistency to eliminate Visual Deviation to guarantee the discrimination and generalization of representations. Considering the complicated interaction between sub-concepts and their visual features, we decompose seen images into visual elements according to their labels and obtain the instance-level sub-deviations from compositions, which is utilized to excavate the category-level primitives of sub-concepts. Furthermore, we present a multi-scale concept composition approach to produce virtual samples from two aspects, which augments the sufficiency and diversity of samples so that the proposed model can generalize to novel compositions. Extensive experiments indicate that our method significantly outperforms the state-of-the- art approaches on three benchmark datasets.
Li et al. (Tue,) studied this question.