Auto data augmentation for testing set

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Testing phase augmentation is a fast way to further improve the performance of image classification when CNN (Convolutional Neural Network) is already trained for hours. Limited attempts have been made to find the best augmentation strategy for testing set. We propose a reinforcement learning based augmentation strategy searching method for testing phase augmentation. With the augmentation strategy, we augment each testing image and integrate features of its augmented images into one feature. The reinforcement learning method searches the best parameters in the augmentation strategy which is formed as a matrix in this paper. Using the proposed method, we achieve competitive accuracies on image classification and face verification.

Original languageEnglish
Title of host publicationPattern Recognition and Computer Vision 2nd Chinese Conference, PRCV 2019, Proceedings, Part II
EditorsZhouchen Lin, Liang Wang, Tieniu Tan, Jian Yang, Guangming Shi, Nanning Zheng, Xilin Chen, Yanning Zhang
PublisherSpringer
Pages66-78
Number of pages13
ISBN (Print)9783030317225
DOIs
StatePublished - 2019
Event2nd Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2019 - Xi'an, China
Duration: 8 Nov 201911 Nov 2019

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11858 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference2nd Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2019
Country/TerritoryChina
CityXi'an
Period8/11/1911/11/19

Keywords

  • Deep reinforcement learning
  • Face verification
  • Image augmentation

Fingerprint

Dive into the research topics of 'Auto data augmentation for testing set'. Together they form a unique fingerprint.

Cite this