Please use this identifier to cite or link to this item:
Scopus Web of Science® Altmetric
Type: Conference paper
Title: Augmentation network for generalised zero-shot learning
Author: Felix Alves, R.
Sasdelli, M.
Reid, I.D.
Carneiro, G.
Citation: Lecture Notes in Artificial Intelligence, 2020 / Ishikawa, H., Liu, C.-L., Pajdla, T., Shi, J. (ed./s), vol.12625, pp.442-458
Publisher: Springer
Publisher Place: Cham, Switzerland
Issue Date: 2020
Series/Report no.: Lecture Notes in Computer Science; 12625
ISBN: 9783030695378
ISSN: 0302-9743
Conference Name: Asian Conference on Computer Vision (ACCV) (30 Nov 2020 - 4 Dec 2020 : virtual online)
Editor: Ishikawa, H.
Liu, C.-L.
Pajdla, T.
Shi, J.
Statement of
Rafael Felix, Michele Sasdelli, Ian Reid, Gustavo Carneiro
Abstract: Generalised zero-shot learning (GZSL) is defined by a training process containing a set of visual samples from seen classes and a set of semantic samples from seen and unseen classes, while the testing process consists of the classification of visual samples from the seen and the unseen classes. Current approaches are based on inference processes that rely on the result of a single modality classifier (visual, semantic, or latent joint space) that balances the classification between the seen and unseen classes using gating mechanisms. There are a couple of problems with such approaches: 1) multi-modal classifiers are known to generally be more accurate than single modality classifiers, and 2) gating mechanisms rely on a complex one-class training of an external domain classifier that modulates the seen and unseen classifiers. In this paper, we mitigate these issues by proposing a novel GZSL method – augmentation network that tackles multi-modal and multi-domain inference for generalised zero-shot learning (AN-GZSL). The multi-modal inference combines visual and semantic classification and automatically balances the seen and unseen classification using temperature calibration, without requiring any gating mechanisms or external domain classifiers. Experiments show that our method produces the new state-of-the-art GZSL results for fine-grained benchmark data sets CUB and FLO and for the large-scale data set ImageNet. We also obtain competitive results for coarse-grained data sets SUN and AWA. We show an ablation study that justifies each stage of the proposed AN-GZSL.
Keywords: Generalised zero-shot learning; Multi-modal inference; Multi-domain inference
Rights: © Springer Nature Switzerland AG 2021
DOI: 10.1007/978-3-030-69538-5_27
Published version:
Appears in Collections:Computer Science publications

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.