Structured Joint Embedding[1] Learning Implemantation
It learns a joint embedding between an input and an output representation. For this example CUB-200-2011[2] Dataset used, ResNet [3] pooling features used as visual features and attributes of classes used as output embeddings.
Reference
[1] - Evaluation of Output Embeddings for Fine-Grained Image Classification, Z. Akata, S. Reed, D. Walter, H.Lee and B.Schiele, CVPR 2015
[2] - Wah C., Branson S., Welinder P., Perona P., Belongie S. “The Caltech-UCSD Birds-200-2011 Dataset.” Computation & Neural Systems Technical Report, CNS-TR-2011-001
[3] - Kaiming He and Xiangyu Zhang and Shaoqing Ren and Jian Sun "Deep Residual Learning for Image Recognition" , arXiv preprint arXiv:1512.03385