Semantic Autoencoder for Zero-Shot Learning
Preprint
Abstract
Existing zero-shot learning (ZSL) models typically learn a projection function from a feature space to a semantic embedding space (e.g.~attribute space). However, such a projection function is only concerned with predicting the training seen class semantic representation (e.g.~attribute prediction) or classification. When applied to test data, which in the context of ZSL contains different (unseen) classes without training data, a ZSL model typically suffers from the project domain shift problem. In this work, we present a novel solution to ZSL based on learning a Semantic AutoEncoder (SAE). Taking the encoder-decoder paradigm, an encoder aims to project a visual feature vector into the semantic space as in the existing ZSL models. However, the decoder exerts an additional constraint, that is, the projection/code must be able to reconstruct the original visual feature. We show that with this additional reconstruction constraint, the learned projection function from the seen classes is able to generalise better to the new unseen classes. Importantly, the encoder and decoder are linear and symmetric which enable us to develop an extremely efficient learning algorithm. Extensive experiments on six benchmark datasets demonstrate that the proposed SAE outperforms significantly the existing ZSL models with the additional benefit of lower computational cost. Furthermore, when the SAE is applied to supervised clustering problem, it also beats the state-of-the-art.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- An Empirical Study and Analysis of Generalized Zero-Shot Learning for Object Recognition in the Wild2016
- Attribute2Image: Conditional Image Generation from Visual Attributes2016
- Learning Deep Representations of Fine-Grained Visual Descriptions2016
- Predicting Deep Zero-Shot Convolutional Neural Networks Using Textual Descriptions2015
- Unsupervised Domain Adaptation for Zero-Shot Learning2015
- Transductive Multi-View Zero-Shot LearningIEEE Transactions on Pattern Analysis and Machine Intelligence, 2015
- Transductive Multi-view Embedding for Zero-Shot Recognition and Annotation2014
- Write a Classifier: Zero-Shot Learning Using Purely Textual Descriptions2013
- Neural networks and principal component analysis: Learning from examples without local minimaNeural Networks, 1989
- Recognition-by-components: A theory of human image understanding.Psychological Review, 1987