Feature Generating Networks for Zero-Shot Learning

Suffering from the extreme training data imbalance between 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 of the art on five challenging datasets — CUB, FLO, SUN, AWA and ImageNet — in both the zero-shot learning and generalized zero-shot learning settings.

@inproceedings{xian2018feature,
    author    = {Xian, Yongqin and Lorenz, Tobias and Schiele, Bernt and Akata, Zeynep},
    title     = {Feature Generating Networks for Zero-Shot Learning},
    year      = {2018},
    month     = {June},
    booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
    pages     = {5542--5551},
    doi       = {10.1109/CVPR.2018.00581}
}