论文标题
学习跨跨性零射击学习的跨域语义 - 视觉关系
Learning Cross-domain Semantic-Visual Relationships for Transductive Zero-Shot Learning
论文作者
论文摘要
零射学习（ZSL）学习识别新课程的模型。 ZSL中的主要挑战之一是训练和测试数据之间的类别不一致引起的域差异。域适应是应对这一挑战的最直观的方式。但是，由于源和目标域之间的不相交标签空间，现有的域适应技术不能直接应用于ZSL。这项工作提出了可转移的语义 - 视觉关系（TSVR）方法，用于转导ZSL。 TSVR重新定义图像识别为预测由类别属性和视觉特征组成的语义 - 视觉融合的相似性/差异标签。在上述转换之后，源和目标域可以具有相同的标签空间，因此可以量化域差异。对于重新定义的问题，相似的语义比对的数量明显小于相似的问题。为此，我们进一步建议使用特定于域的批准归一化来对齐域差异。
Zero-Shot Learning (ZSL) learns models for recognizing new classes. One of the main challenges in ZSL is the domain discrepancy caused by the category inconsistency between training and testing data. Domain adaptation is the most intuitive way to address this challenge. However, existing domain adaptation techniques cannot be directly applied into ZSL due to the disjoint label space between source and target domains. This work proposes the Transferrable Semantic-Visual Relation (TSVR) approach towards transductive ZSL. TSVR redefines image recognition as predicting the similarity/dissimilarity labels for semantic-visual fusions consisting of class attributes and visual features. After the above transformation, the source and target domains can have the same label space, which hence enables to quantify domain discrepancy. For the redefined problem, the number of similar semantic-visual pairs is significantly smaller than that of dissimilar ones. To this end, we further propose to use Domain-Specific Batch Normalization to align the domain discrepancy.
