Generalized zero-shot classification is predicting the labels of the test images coming from seen or unseen classes. The task is difficult because of the bias problem, that is, unseen samples are easily to be misclassified to seen classes. Many methods have handled the problem by training a generative adversarial network (GAN) to generate fake samples. However, the GAN model trained with seen samples might not be appropriate for generating unseen samples. For dealing with this problem, we learn a bias alleviating generative adversarial network for generalized zero-shot classification by generating seen and unseen samples, simultaneously. We train the generator to generate more realistic unseen samples by adding semantic similarity and cluster center regularizations to alleviate the bias problem. The semantic similarity regularization is to restrict the relationships of the generated unseen visual prototypes and seen visual prototypes by their class prototypes to avoid the generated unseen samples similar to the seen samples. The cluster center regularization is to utilize the cluster property of target data to make the generated unseen visual prototypes near to the most similar cluster centers, generating realistic unseen samples. From the experiments, we can see the proposed method achieves promising results.

广义零镜头分类可以预测来自可见或不可见类别的测试图像标签。由于存在偏见问题，因此该任务很困难，也就是说，看不见的样本很容易被误分类为可见的类别。许多方法通过训练生成对抗网络（GAN）来生成假样本来解决该问题。但是，用可见样本训练的GAN模型可能不适用于生成看不见的样本。为了解决这个问题，我们通过同时生成可见和不可见样本，学习了一种用于广义零镜头分类的减轻偏向的生成对抗网络。我们通过添加语义相似性和聚类中心正则化来训练生成器生成更现实的，看不见的样本，以缓解偏差问题。语义相似性正则化是通过它们的类原型来限制所生成的看不见的视觉原型和所见的视觉原型之间的关系，以避免所生成的与所见的样本相似的看不见的样本。聚类中心正则化是利用目标数据的聚类属性使生成的看不见的视觉原型靠近最相似的聚类中心，从而生成逼真的看不见的样本。从实验中可以看出，该方法取得了很好的效果。聚类中心正则化是利用目标数据的聚类属性使生成的看不见的视觉原型靠近最相似的聚类中心，从而生成逼真的看不见的样本。从实验中可以看出，该方法取得了很好的效果。聚类中心正则化是利用目标数据的聚类属性使生成的看不见的视觉原型靠近最相似的聚类中心，从而生成逼真的看不见的样本。从实验中可以看出，该方法取得了很好的效果。