Learning Visually Consistent Label Embeddings for Zero-Shot Learning

Demirel, Berkan; Cinbiş, Ramazan Gökberk; Ikizler-Cinbis, Nazli

doi:10.1109/icip.2019.8803458

Cited by 9 publications

(11 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To verify the validity of the model, we set up the different cases for testing: (a) the testing set contains only seen instances (s); (b) the testing set contains only unseen instances (u); (c) both seen and unseen instances (s+u) are present in the testing set. It is worth noting that the mAP of our method in the seen class improves from 65.6% to 71.1% compared with Demirel 49 . In the ZSD task, our method outperforms the mAP of Demirel 49 and SPGP 7 by 13.6% and 3.3%, respectively.…”

Section: Resultsmentioning

confidence: 81%

Adaptive adjustment with semantic embedding for zero-shot object detection

Shi

Tan

et al. 2023

J. Electron. Imag.

View full text Add to dashboard Cite

.Traditional zero-shot object-detection algorithms detect images of untrained classes in the model with the help of semantic embedding. However, these approaches may perform poorly due to the limitations of fixed semantic embedding. Given that fixed semantic attributes lead to a lack of generalization capabilities in the model, a semantic enhancement mechanism is proposed to update the semantic embedding, which is used to serve the needs of the visual space. Specifically, considering that the original semantic space is not enough to construct a visual-semantic mapping relationship, an augmented semantic embedding (ASE) approach is designed to supplement semantic attribute information. Then, a semantic channel attention mechanism is used to adjust the ASE. The adjustment strategy retains adequate attribute information, which is highly relevant to visual features. Finally, to alleviate the domain shift problem, a clustering association strategy is introduced to establish an inferred relationship, which ensures that the predictor is generalized to the unseen domain during training. The superiority of the proposed method is demonstrated by the MS-COCO and PASCAL VOC datasets.

show abstract

Section: Resultsmentioning

confidence: 81%

Adaptive adjustment with semantic embedding for zero-shot object detection

Shi

Tan

et al. 2023

J. Electron. Imag.

View full text Add to dashboard Cite

show abstract

“…One common point of the above zero‐shot learning can be concluded as they learn a projection from the training data (seen classes) to the semantic embedding space. Inspired by this, zero‐shot learning is applied to object recognition and detection . Demirel et al .…”

Section: Related Workmentioning

confidence: 99%

“…Demirel et al . propose a hybrid region embedding for ZSD that combines two mainstream embedding approaches in zero‐shot learning. Bansal et al .…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

TemplateFree: product detection on retail store shelves

Sun

Zhang

Akashi

2019

IEEJ Transactions Elec Engng

View full text Add to dashboard Cite

Product recognition has a significant role because of its benefits to the compliant arrangements of stores, which further affects commercial contracts, customer satisfaction, and sale achievement. Automatic recognition systems have been proposed owing to the current high cost of manual inspection by clerks. Because of the difficult collection of product images, the systems are commonly one-shot cases, in which the training data are actually template product images. However, despite the development of one-shot recognition, the systems rarely utilize the special characteristics of products on retail store shelves, and the frequent updating of templates is still challenging. Furthermore, we consider that product detection can be the basis of product recognition. In this article, instead of the present workflow, we propose a novel product detection system, named TemplateFree, which combines product segmentation and zero-shot learning. It detects products on retail store shelves by single store shelf images, that is, corresponding template product images are not necessary. TemplateFree concentrates on the characteristic that a store shelf can be segmented horizontally into layers and then vertically into products so that each product can be detected according to the segmentation. Double zero-shot deep learning frameworks are used to improve the segmentation. In experiments, TemplateFree achieves better results than the present method.

show abstract

“…ZSL refers to the technology of using some known category data and the auxiliary information corresponding to the known category to train a certain model, so as to realize the classification and recognition of the data of the unknown category. A ZSL approach is proposed to simulate knowledge transfer between classes by learning visually consistent word vectors and tag embedding models (Demirel et al, 2019). The main idea is to project the vector space word vectors of attributes and classes into the visual space, so as to make the word representation of semantically related classes more close and, furthermore, use the proposed projection vector embedded in the model to identify the invisible classes.…”

Section: Introductionmentioning

confidence: 99%

Nonintrusive Monitoring for Electric Vehicles Based on Zero-Shot Learning

Ren

et al. 2021

Front. Energy Res.

View full text Add to dashboard Cite

Monitoring the charging behavior of electric vehicle clusters will contribute to developing more effective energy management strategies for grid operators. A low implementation cost leads to a wide application prospect in nonintrusive monitoring for EVs. Aiming at the problem that traditional nonintrusive monitoring methods cannot identify unknown devices accurately due to the lack of classes, a nonintrusive monitoring method based on zero-shot learning (ZSL) is proposed in this article, one which can monitor the unknown types of EVs connected to charging piles. First, the charging characteristics of known EVs and unknown EVs are extracted by dictionary learning. Then EVs are classified by ZSL based on sparse coding. Furthermore, EVs are decomposed based on the proposed multimode factorial hidden Markov model (FHMM). Finally, the EV dataset of Pecan Street is used to verify the effectiveness and accuracy of the proposed method.

show abstract

Learning Visually Consistent Label Embeddings for Zero-Shot Learning

Cited by 9 publications

References 33 publications

Adaptive adjustment with semantic embedding for zero-shot object detection

Adaptive adjustment with semantic embedding for zero-shot object detection

TemplateFree: product detection on retail store shelves

Nonintrusive Monitoring for Electric Vehicles Based on Zero-Shot Learning

Contact Info

Product

Resources

About