Triple Verification Network for Generalized Zero-Shot Learning

Zhang, Haofeng; Long, Yang; Guan, Yu; Shao, Ling

doi:10.1109/tip.2018.2869696

Cited by 85 publications

(50 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2. Concretely, besides the baseline results recorded in [25], we also provide the results of four other methods, including TVN [48], VZSL [49], LESAE [13] and LESD [11], among which LESAE and LESD are low rank based methods and most related to ours. From Tab.…”

Section: Results On Zslmentioning

confidence: 99%

Prototype Relaxation With Robust Principal Component Analysis for Zero Shot Learning

Mao

Zhang

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

Zero Shot Learning (ZSL) has been attracting increasing attention due to its powerful ability of recognizing objects of unseen classes. As one type of ZSL methods, the low rank based strategy has achieved remarkable success. However, traditional low rank based methods are often based on the assumption that a variety of visual features from a same class can be projected to a single attribute by ignoring the background information and other noisy interference in visual features. This assumption is unreasonable and often leads to bad performance when there is big variance within a class. In this paper, a novel method called Prototype Relaxation with Robust Principal Component Analysis (RPCA) is proposed to relax this assumption by adding a sparse noise constraint. In addition, to avoid the confusion between similar classes, an orthogonal constraint is employed to disperse all the class prototypes, including both seen and unseen classes, in latent space. Furthermore, to alleviate the domain shift problem, vectors from latent space are exploited to reconstruct visual features and semantic attributes respectively. Besides, the hubness problem is also mitigated by applying the max probability model in all three spaces. Extensive experiments are conducted on four popular datasets and the results demonstrate the superiority of this method.

show abstract

Section: Results On Zslmentioning

confidence: 99%

Prototype Relaxation With Robust Principal Component Analysis for Zero Shot Learning

Mao

Zhang

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recent work showed that generating synthetic samples of unseen classes using GANs or VAEs [48,11,5,57] can substantially improve generalized zero-shot learning . The recent literature considers this generative effort to be orthogonal to modelling, since the two efforts can be combined [29,7,56,15,54]. Here we compare COSMO directly both with the approaches listed above, and with generative approaches fCLSWGAN [48], cycle-(U)WGAN [11], SE-GZSL [5].…”

Section: Compared Methodsmentioning

confidence: 99%

Adaptive Confidence Smoothing for Generalized Zero-Shot Learning

Atzmon

Chechik

2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

105

View full text Add to dashboard Cite

Generalized zero-shot learning (GZSL) is the problem of learning a classifier where some classes have samples and others are learned from side information, like semantic attributes or text description, in a zero-shot learning fashion (ZSL). Training a single model that operates in these two regimes simultaneously is challenging. Here we describe a probabilistic approach that breaks the model into three modular components, and then combines them in a consistent way. Specifically, our model consists of three classifiers: A "gating" model that makes soft decisions if a sample is from a "seen" class, and two experts: a ZSL expert, and an expert model for seen classes. We address two main difficulties in this approach: How to provide an accurate estimate of the gating probability without any training samples for unseen classes; and how to use expert predictions when it observes samples outside of its domain.The key insight to our approach is to pass information between the three models to improve each one's accuracy, while maintaining the modular structure. We test our approach, adaptive confidence smoothing (COSMO ), on four standard GZSL benchmark datasets and find that it largely outperforms state-of-the-art GZSL models. COSMO is also the first model that closes the gap and surpasses the performance of generative models for GZSL, even-though it is a light-weight model that is much easier to train and tune. Notably, COSMO offers a new view for developing zeroshot models. Thanks to COSMO's modular structure, instead of trying to perform well both on seen and on unseen classes, models can focus on accurate classification of unseen classes, and later consider seen class models.

show abstract

“…Therefore, they propose the new task-Generalised ZSL, which assumes that the search space of the nearest neighbour should be extended to both seen and unseen classes. Recently, Zhang et al in [35] considered GZSL problem as a triple verification problem and a novel optimization of regression and compatibility function is proposed to solve this problem. Subsequently, Xian et al [36] put forward a new standard split of several popular datasets for GZSL testing, and release a benchmark of some recent ZSL methods, which makes the later researchers more convenient and has greatly promoted the development of ZSL.…”

Section: Generalized Zero Shot Learning (Gzsl)mentioning

confidence: 99%

Semantic combined network for zero‐shot scene parsing

Wang

Zhang

Wang

et al. 2020

IET Image Processing

Self Cite

View full text Add to dashboard Cite

2020) 'Semantic combined network for zero-shot scene parsing . ', IET image processing., 14 (4). 757 -765. The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-prot purposes provided that:• a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders.Please consult the full DRO policy for further details.Abstract: Recently, image-based scene parsing has attracted increasing attention due to its wide application. However, conventional models can only be valid on images with the same domain of the training set, and are typically trained using discrete and meaningless labels. Inspired by the traditional zero shot learning methods which employ an auxiliary side information to bridge the source and target domains, we propose a novel framework called Semantic Combined Network (SCN), which aims at learning a scene parsing model only from the images of the seen classes while targeting on the unseen ones. In addition, with the assist of semantic embeddings of classes, our SCN can further improve the performances of traditional fully supervised scene parsing methods. Extensive experiments are conducted on the dataset Cityscapes, and the results show that our SCN can perform well on both Zero Shot Scene Parsing (ZSSP) and Generalized ZSSP (GZSSP) settings based on several state-of-the-art scene parsing architectures. Furthermore, we test our model under the traditional fully supervised setting and the results show that our SCN can also significantly improve the performances of the original network models.

show abstract

Triple Verification Network for Generalized Zero-Shot Learning

Cited by 85 publications

References 25 publications

Prototype Relaxation With Robust Principal Component Analysis for Zero Shot Learning

Prototype Relaxation With Robust Principal Component Analysis for Zero Shot Learning

Adaptive Confidence Smoothing for Generalized Zero-Shot Learning

Semantic combined network for zero‐shot scene parsing

Contact Info

Product

Resources

About