Training and testing labelled image and video datasets of human faces for different indoor visual comfort and glare visual discomfort situations

Hicham Johra; Martin Veit; Mathias Østergaard Poulsen; Albert Daugbjerg Christensen; Rikke Gade; Thomas B. Moeslund; Rasmus Lund Jensen

doi:10.54337/aau542153983

Training and testing labelled image and video datasets of human faces for different indoor visual comfort and glare visual discomfort situations

Hicham Johra^*, Martin Veit, Mathias Østergaard Poulsen, Albert Daugbjerg Christensen, Rikke Gade, Thomas B. Moeslund, Rasmus Lund Jensen

^*Corresponding author for this work

Research output: Book/Report › Report › Research

99 Downloads (Pure)

Abstract

The aim of this technical report is to provide a description and access to labelled image and video datasets of human faces that have been generated for different indoor visual comfort and glare visual discomfort situations. These datasets have been used to train and test a computer-vision artificial neural network detecting glare discomfort from images of human faces.

Original language	English

Place of Publication	Aalborg
Publisher	Department of the Built Environment, Aalborg University
Number of pages	20
DOIs	https://doi.org/10.54337/aau542153983
Publication status	Published - Jul 2023

Series	DCE Technical Reports
Number	316
ISSN	1901-726X

Keywords

glare
glare discomfort
visual discomfort
empirical data
human face analysis
AI
CNN
Computer Vision
training dataset

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.54337/aau542153983

Tech Report 316 - SCODYF training datasetFinal published version, 493 KBLicence: CC BY-NC 4.0

AUB Link

Search for the material in Aalborg University Library's search engine

1 Finished

AI for optimizing indoor visual comfort from facial analysis
Johra, H. (PI), Moeslund, T. B. (Project Participant), Gade, R. (Project Participant), Jensen, R. L. (Project Participant), Petrova, E. A. (Project Participant) & Rohde, L. (Project Participant)
01/01/2021 → 30/06/2021
Project: Research

1 Conference article in Journal

Artificial Intelligence for Detecting Indoor Visual Discomfort from Facial Analysis of Building Occupants
Johra, H., Gade, R., Poulsen, M. Ø., Christensen, A. D., Sarey Khanie, M., Moeslund, T. B. & Jensen, R. L., 2021, In: Journal of Physics: Conference Series (Online). 2042, 1, 012008.
Research output: Contribution to journal › Conference article in Journal › Research › peer-review

Open Access
File
5 Citations (Scopus)

88 Downloads (Pure)

Training and testing labelled dataset of human faces for different indoor visual comfort and glare visual discomfort situations - images
Johra, H. (Creator) & Veit, M. (Creator), Zenodo, 1 Jun 2022
DOI: 10.5281/zenodo.6895080, https://zenodo.org/record/6895080
Dataset
Training and testing labelled dataset of human faces for different indoor visual comfort and glare visual discomfort situations - videos
Johra, H. (Creator) & Veit, M. (Creator), Zenodo, 1 Jun 2022
DOI: 10.5281/zenodo.6895054, https://zenodo.org/record/6895054
Dataset

Cite this

Johra, H., Veit, M., Poulsen, M. Ø., Christensen, A. D., Gade, R., Moeslund, T. B., & Jensen, R. L. (2023). Training and testing labelled image and video datasets of human faces for different indoor visual comfort and glare visual discomfort situations. Department of the Built Environment, Aalborg University. DCE Technical Reports No. 316 https://doi.org/10.54337/aau542153983

@book{414dbd6ba3ba4026935d2a8c1bddb967,

title = "Training and testing labelled image and video datasets of human faces for different indoor visual comfort and glare visual discomfort situations",

abstract = "The aim of this technical report is to provide a description and access to labelled image and video datasets of human faces that have been generated for different indoor visual comfort and glare visual discomfort situations. These datasets have been used to train and test a computer-vision artificial neural network detecting glare discomfort from images of human faces.",

keywords = "glare, glare discomfort, visual discomfort, empirical data, human face analysis, AI, CNN, Computer Vision, training dataset, Aalborg University, building study case",

author = "Hicham Johra and Martin Veit and Poulsen, {Mathias {\O}stergaard} and Christensen, {Albert Daugbjerg} and Rikke Gade and Moeslund, {Thomas B.} and Jensen, {Rasmus Lund}",

year = "2023",

month = jul,

doi = "10.54337/aau542153983",

language = "English",

series = "DCE Technical Reports",

number = "316",

publisher = "Department of the Built Environment, Aalborg University",

}

Johra, H , Veit, M, Poulsen, MØ, Christensen, AD, Gade, R , Moeslund, TB & Jensen, RL 2023, Training and testing labelled image and video datasets of human faces for different indoor visual comfort and glare visual discomfort situations. DCE Technical Reports, no. 316, Department of the Built Environment, Aalborg University, Aalborg. https://doi.org/10.54337/aau542153983

Training and testing labelled image and video datasets of human faces for different indoor visual comfort and glare visual discomfort situations. / Johra, Hicham ; Veit, Martin; Poulsen, Mathias Østergaard et al.
Aalborg: Department of the Built Environment, Aalborg University, 2023. 20 p. (DCE Technical Reports; No. 316).

Research output: Book/Report › Report › Research

TY - RPRT

T1 - Training and testing labelled image and video datasets of human faces for different indoor visual comfort and glare visual discomfort situations

AU - Johra, Hicham

AU - Veit, Martin

AU - Poulsen, Mathias Østergaard

AU - Christensen, Albert Daugbjerg

AU - Gade, Rikke

AU - Moeslund, Thomas B.

AU - Jensen, Rasmus Lund

PY - 2023/7

Y1 - 2023/7

N2 - The aim of this technical report is to provide a description and access to labelled image and video datasets of human faces that have been generated for different indoor visual comfort and glare visual discomfort situations. These datasets have been used to train and test a computer-vision artificial neural network detecting glare discomfort from images of human faces.

AB - The aim of this technical report is to provide a description and access to labelled image and video datasets of human faces that have been generated for different indoor visual comfort and glare visual discomfort situations. These datasets have been used to train and test a computer-vision artificial neural network detecting glare discomfort from images of human faces.

KW - glare

KW - glare discomfort

KW - visual discomfort

KW - empirical data

KW - human face analysis

KW - AI

KW - CNN

KW - Computer Vision

KW - training dataset

KW - Aalborg University

KW - building study case

U2 - 10.54337/aau542153983

DO - 10.54337/aau542153983

M3 - Report

T3 - DCE Technical Reports

BT - Training and testing labelled image and video datasets of human faces for different indoor visual comfort and glare visual discomfort situations

PB - Department of the Built Environment, Aalborg University

CY - Aalborg

ER -

Training and testing labelled image and video datasets of human faces for different indoor visual comfort and glare visual discomfort situations

Abstract

Keywords

UN SDGs

Access to Document

AUB Link

Fingerprint

Projects

AI for optimizing indoor visual comfort from facial analysis

Research output

Artificial Intelligence for Detecting Indoor Visual Discomfort from Facial Analysis of Building Occupants

Datasets

Training and testing labelled dataset of human faces for different indoor visual comfort and glare visual discomfort situations - images

Training and testing labelled dataset of human faces for different indoor visual comfort and glare visual discomfort situations - videos

Cite this