Deep feature representation based on privileged knowledge transfer

•A novel feature learning frame using knowledge transfer is proposed.•Training the PKT-network guided by privileged knowledge and applying to multi-object image retrieval.•Achieving outstanding performance using the PKT-network on two datasets. Feature representation has been an elusive concept unti...

Full description

Saved in:
Bibliographic Details
Published in:Pattern recognition letters 2019-03, Vol.119, p.62-70
Main Authors: Duan, Lijuan, En, Qing, Qiao, Yuanhua, Cui, Song, Qing, Laiyun
Format: Article
Language:English
Subjects:
Deep neural network
Distillation
Feature representation
Knowledge management
Knowledge representation
Labels
Learning
Lower bounds
Multi-object retrieval
Neural networks
Privileged knowledge transfer
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•A novel feature learning frame using knowledge transfer is proposed.•Training the PKT-network guided by privileged knowledge and applying to multi-object image retrieval.•Achieving outstanding performance using the PKT-network on two datasets. Feature representation has been an elusive concept until neural networks become popular and exhibit the strong learning capability. However, as supervised labels provide a little meaningful information for feature representation, neural networks suffer from inadequate ability to acquire their representing knowledge by extracting patterns from raw data. To address this issue, in this paper, we propose a novel feature learning model by transferring the spatial relation information to the weights of neural networks. More specially, privileged knowledge stemmed from segmented image is fully utilized in distilling stage to perceive notable objects. The segmented image is regarded as a probability distribution of objects, the notable objects are fetched by maximizing lower bound of mutual information, instead of considering each image as an example of one-hot label. Through this way, spatial relation information is contributed to the feature learning besides class labels. This strategy, denoted as PKT-network, is applied to the image network training. Experimental results show that PKT-network performs excellently for multi-objects representation on Pascal VOC 2012 SegmentationClass (20 object classes) dataset and Microsoft COCO (80 object classes) dataset.
ISSN:0167-8655
1872-7344
DOI:10.1016/j.patrec.2017.12.004