KomNET: Face Image Dataset from Various Media for Face Recognition

I Nyoman Gede Arya Astawa,I Ketut Gede Darma Putra,Made Sudarma,Rukmi Sari Hartati

doi:10.1016/j.dib.2020.105677

I Nyoman Gede Arya Astawa, I Ketut Gede Darma Putra + Show 2 more

Open Access

https://doi.org/10.1016/j.dib.2020.105677

Copy DOI

Journal: Data in Brief	Publication Date: May 13, 2020
Citations: 5	License type: cc-by

Affiliation: Jakarta State Polytechnic, Udayana University

Abstract

KomNet is a face image dataset originated from three media sources which can be used to recognize faces. KomNET contains face images which were collected from three different media sources, i.e. mobile phone camera, digital camera, and media social. The collected face dataset was frontal face image or facing the camera. The face dataset originated from the three media were collected without certain conditions such as lighting, background, haircut, mustache and beard, head cover, glasses, and differences of expression. KomNet dataset were collected from 50 clusters in which each of them consisted of 24 face images. To increase the number of training data, the face images were propagated with augmentation image technique, in which ten augmentations were used such as Rotate, Flip, Gaussian Blur, Gamma Contrast, Sigmoid Contrast, Sharpen, Emboss, Histogram Equalization, Hue and Saturation, Average Blur so the face images became 240 face images per cluster. The author trained the dataset by using CNN-based transfer learning VGGface. KomNET dataset are freely available on https://data.mendeley.com/datasets/hsv83m5zbb/2.

Highlights

KomNet is a face image dataset originated from three media sources which can be used to recognize faces
To increase the number of training data, the face images were propagated with augmentation image technique, in which ten augmentations were used such as Rotate, Flip, Gaussian Blur, Gamma Contrast, Sigmoid Contrast, Sharpen, Emboss, Histogram Equalization, Hue and Saturation, Average Blur so the face images became 240 face images per cluster
Raw digital image (.jpg, .jpeg, .png) Filtered augmentation image: average blur, emboss, flip, gamma contrast, gaussian blur, histogram equalization, rotate, hue and saturation, sharpen, and sigmoid contrast (.jpg, .jpeg, .png) the collected face images that were collected from three different sources were frontal face image or facing the camera Face images were collected from three different sources

Summary

Introduction

KomNET: Face Image Dataset from Various Media for Face Recognition KomNet is a face image dataset originated from three media sources which can be used to recognize faces. KomNET contains face images which were collected from three different media sources, i.e. mobile phone camera, digital camera, and media social. The face dataset originated from the three media were collected without certain conditions such as lighting, background, haircut, mustache and beard, head cover, glasses, and differences of expression.

Results

Conclusion