Bai nationality has a long history and has its own language. Limited by the fact that there are fewer and fewer people who know the Bai language, the literature and culture of the Bai nationality begin to lose rapidly. In order to make the people who do not understand Bai characters can also read the ancient books of Bai nationality, this paper is based on the research of high-precision single character recognition model of Bai characters. First, with the help of Bai culture lovers and related scholars, we have constructed a data set of Bai characters, but limited by the need of expert knowledge, so the data set is limited in size. As a result, deep learning models with the nature of data hunger cannot get an ideal accuracy. In order to solve this issue, we propose to use the Chinese data set which also belongs to Sino-Tibetan language family to improve the recognition accuracy of Bai characters through transfer learning. In addition, we propose four transfer learning approaches: Direct Knowledge Transfer (DKT), Indirect Knowledge Transfer (IKT), Self-coding Knowledge Transfer (SCKT), and Self-supervised Knowledge Transfer (SSKT). Experiments show that our approaches greatly improve the recognition accuracy of Bai characters.

Conventional zero-shot learning aims to train a classifier on a training set (seen classes) to recognize instances of novel classes (unseen classes) by class-level semantic attributes. In generalized zero-shot learning (GZSL), the classifier needs to recognize both seen and unseen classes, which is a problem of extreme data imbalance. To solve this problem, feature generative methods have been proposed to make up for the lack of unseen classes. Current generative methods use class semantic attributes as the cues for synthetic visual features, which can be considered mapping of the semantic attribute to visual features. However, this mapping cannot effectively transfer knowledge learned from seen classes to unseen classes because the information in the semantic attributes and the information in visual features are asymmetric: semantic attributes contain key category description information, while visual features consist of visual information that cannot be represented by semantics. To this end, we propose a residual-prototype-generating network (RPGN) for GZSL that extracts the residual visual features from original visual features by an encoder–decoder and synthesizes the prototype visual features associated with semantic attributes by a disentangle regressor. Experimental results show that the proposed method achieves competitive results on four GZSL benchmark datasets with significant gains.

No abstract

Unlike conventional zero-shot learning (CZSL) which only focuses on the recognition of unseen classes by using the classifier trained on seen classes and semantic embeddings, generalized zero-shot learning (GZSL) aims at recognizing both the seen and unseen classes, so it is more challenging due to the extreme training imbalance. Recently, some feature generation methods introduce metric learning to enhance the discriminability of visual features. Although these methods achieve good results, they focus only on metric learning in the visual feature space to enhance features and ignore the association between the feature space and the semantic space. Since the GZSL method uses semantics as prior knowledge to migrate visual knowledge to unseen classes, the consistency between visual space and semantic space is critical. To this end, we propose relational metric learning which can relate the metrics in the two spaces and make the distribution of the two spaces more consistent. Based on the generation method and relational metric learning, we proposed a novel GZSL method, termed VS-Boost, which can effectively boost the association between vision and semantics. The experimental results demonstrate that our method is effective and achieves significant gains on five benchmark datasets compared with the state-of-the-art methods.