Attributes2Classname: A discriminative model for attribute-based unsupervised zero-shot learning

05/04/2017
by   Berkan Demirel, et al.
0

We propose a novel approach for unsupervised zero-shot learning (ZSL) of classes based on their names. Most existing unsupervised ZSL methods aim to learn a model for directly comparing image features and class names. However, this proves to be a difficult task due to dominance of non-visual semantics in underlying vector-space embeddings of class names. To address this issue, we discriminatively learn a word representation such that the similarities between class and combination of attribute names fall in line with the visual similarity. Contrary to the traditional zero-shot learning approaches that are built upon attribute presence, our approach bypasses the laborious attribute-class relation annotations for unseen classes. In addition, our proposed approach renders text-only training possible, hence, the training can be augmented without the need to collect additional image data. The experimental results show that our method yields state-of-the-art results for unsupervised ZSL in three benchmark datasets.

READ FULL TEXT
research
08/19/2020

Attribute Prototype Network for Zero-Shot Learning

From the beginning of zero-shot learning research, visual attributes hav...
research
10/15/2016

Recovering the Missing Link: Predicting Class-Attribute Associations for Unsupervised Zero-Shot Learning

Collecting training images for all visual categories is not only expensi...
research
02/05/2018

Zero-Shot Kernel Learning

In this paper, we address an open problem of zero-shot learning. Its pri...
research
09/06/2021

Zero-Shot Open Set Detection by Extending CLIP

In a regular open set detection problem, samples of known classes (also ...
research
03/02/2020

Addressing target shift in zero-shot learning using grouped adversarial learning

In this paper, we present a new paradigm to zero-shot learning (ZSL) tha...
research
09/12/2017

Joint Dictionaries for Zero-Shot Learning

A classic approach toward zero-shot learning (ZSL) is to map the input d...
research
12/11/2018

Zero-Shot Learning with Sparse Attribute Propagation

Zero-shot learning (ZSL) aims to recognize a set of unseen classes witho...

Please sign up or login with your details

Forgot password? Click here to reset