Suffering from the extreme training data imbalance be-tween seen and unseen classes, most of existing state-of-the-art approaches fail to achieve satisfactory results for the challenging generalized zero-shot learning task. To circumvent the need for labeled examples of unseen classes, we propose a novel generative adversarial network (GAN) that synthesizes CNN features conditioned on class-level semantic information, offering a shortcut directly from a semantic descriptor of a class to a class-conditional feature distribution. Our proposed approach, pairing a Wasserstein GAN with a classification loss, is able to generate sufficiently discriminative CNN features to train softmax classifiers or any multimodal embedding method. Our experimental results demonstrate a significant boost in accuracy over the state ofthe art on five challenging datasets – CUB, FLO, SUN, AWA and ImageNet – in both the zero-shot learning and generalized zero-shot learning settings.
IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
2018-06-19
2021-08-08 19:51:00