Synthesized Classifiers for Zero-Shot Learning
Authors:
- Soravit ChangpinyoSCSoravit Changpinyo
- Wei-Lun ChaoWCWei-Lun Chao
- Boqing GongBGBoqing Gong
- Fei ShaFSFei Sha
Top Cited Papers
- 1 June 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 5327-5336
Abstract
Given semantic descriptions of object classes, zero-shot learning aims to accurately recognize objects of the unseen classes, from which no examples are available at the training stage, by associating them to the seen classes, from which labeled examples are provided. We propose to tackle this problem from the perspective of manifold learning. Our main idea is to align the semantic space that is derived from external information to the model space that concerns itself with recognizing visual features. To this end, we introduce a set of "phantom" object classes whose coordinates live in both the semantic space and the model space. Serving as bases in a dictionary, they can be optimized from labeled data such that the synthesized real object classifiers achieve optimal discriminative performance. We demonstrate superior accuracy of our approach over the state of the art on four benchmark datasets for zero-shot learning, including the full ImageNet Fall 2011 dataset with more than 20,000 unseen classes.Keywords
All Related Versions
This publication has 27 references indexed in Scilit:
- Active Transfer Learning with Zero-Shot Priors: Reusing Past Datasets for Future Tasks2015
- Zero-shot object recognition by semantic manifold distance2015
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- Transductive Multi-View Zero-Shot LearningIEEE Transactions on Pattern Analysis and Machine Intelligence, 2015
- Caffe2014
- Decorrelating Semantic Visual Attributes by Resisting the Urge to Share2014
- The SUN Attribute Database: Beyond Categories for Deeper Scene UnderstandingInternational Journal of Computer Vision, 2014
- Relative attributes2011
- Evaluating knowledge transfer and zero-shot learning in a large-scale setting2011
- What helps where – and why? Semantic relatedness for knowledge transfer2010