Relation-based Discriminative Cooperation Network for Zero-Shot Classification
نویسندگان
چکیده
• The discriminative visual embedding preserves the information of image features by separating inter-classes and clustering intra-classes with a margin. semantic acts as pivot regularization to ensure cooperated structures representative utilizing relations between classes. Extensive experimental evaluation on multiple datasets, including large scale ImageNet shows that proposed model performs favorably against state-of-the-art ZSL methods. Zero-shot learning (ZSL) aims assign category corresponding relevant label unseen sample based relationship learned features. However, most typical models faced domain bias problem, which leads or test samples being easily misclassified into seen training categories. To handle this we propose relation-based cooperation network (RDCN) for in work. effectively utilize robust metric space spanned semantics help set relations. On other hand, devise relation measure embedded semantics, validation will guide module learn more information. At last, RDCN is validated six benchmarks, extensive experiments demonstrate superiority method over existing traditional zero-shot setting realistic generalized setting.
منابع مشابه
Hard Negative Mining for Metric Learning Based Zero-Shot Classification
Zero-Shot learning has been shown to be an efficient strategy for domain adaptation. In this context, this paper builds on the recent work of Bucher et al. [1], which proposed an approach to solve ZeroShot classification problems (ZSC) by introducing a novel metric learning based objective function. This objective function allows to learn an optimal embedding of the attributes jointly with a me...
متن کاملDiscriminative Learning of Latent Features for Zero-Shot Recognition
Zero-shot learning (ZSL) aims to recognize unseen image categories by learning an embedding space between image and semantic representations. For years, among existing works, it has been the center task to learn the proper mapping matrices aligning the visual and semantic space, whilst the importance to learn discriminative representations for ZSL is ignored. In this work, we retrospect existin...
متن کاملZero-Shot Learning for Semantic Utterance Classification
We propose a novel zero-shot learning method for semantic utterance classification (SUC). It learns a classifier f : X → Y for problems where none of the semantic categories Y are present in the training set. The framework uncovers the link between categories and utterances through a semantic space. We show that this semantic space can be learned by deep neural networks trained on large amounts...
متن کاملZero-Shot Relation Extraction via Reading Comprehension
We show that relation extraction can be reduced to answering simple reading comprehension questions, by associating one or more natural-language questions with each relation slot. This reduction has several advantages: we can (1) learn relationextraction models by extending recent neural reading-comprehension techniques, (2) build very large training sets for those models by combining relation-...
متن کاملProbabilistic Zero-shot Classification with Semantic Rankings
In this paper we propose a non-metric rankingbased representation of semantic similarity that allows natural aggregation of semantic information from multiple heterogeneous sources. We apply the ranking-based representation to zeroshot learning problems, and present deterministic and probabilistic zero-shot classifiers which can be built from pre-trained classifiers without retraining. We demon...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Pattern Recognition
سال: 2021
ISSN: ['1873-5142', '0031-3203']
DOI: https://doi.org/10.1016/j.patcog.2021.108024