An EEG-based image annotation system

Viral Parekh, Ramanathan Subramanian, Dipanjan Roy, C. V. Jawahar

Research output: A Conference proceeding or a Chapter in BookOther chapter contributionpeer-review

16 Citations (Scopus)

Abstract

The success of deep learning in computer vision has greatly increased the need for annotated image datasets. We propose an EEG (Electroencephalogram)-based image annotation system. While humans can recognize objects in 20–200 ms, the need to manually label images results in a low annotation throughput. Our system employs brain signals captured via a consumer EEG device to achieve an annotation rate of up to 10 images per second. We exploit the P300 event-related potential (ERP) signature to identify target images during a rapid serial visual presentation (RSVP) task. We further perform unsupervised outlier removal to achieve an F1-score of 0.88 on the test set. The proposed system does not depend on category-specific EEG signatures enabling the annotation of any new image category without any model pre-training.

Original languageEnglish
Title of host publicationComputer Vision, Pattern Recognition, Image Processing, and Graphics - 6th National Conference, NCVPRIPG 2017, Revised Selected Papers
EditorsRenu Rameshan, Sumantra Dutta Roy, Chetan Arora
PublisherSpringer
Pages303-313
Number of pages11
VolumeNetherlands
ISBN (Electronic)9789811300202
ISBN (Print)9789811300196
DOIs
Publication statusPublished - 2018
Externally publishedYes
Event6th National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics, NCVPRIPG 2017 - Mandi, India
Duration: 16 Dec 201719 Dec 2017

Publication series

NameCommunications in Computer and Information Science
Volume841
ISSN (Print)1865-0929

Conference

Conference6th National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics, NCVPRIPG 2017
Country/TerritoryIndia
CityMandi
Period16/12/1719/12/17

Fingerprint

Dive into the research topics of 'An EEG-based image annotation system'. Together they form a unique fingerprint.

Cite this