CoChat

Recognizing Human Races through Machine Learning—A Multi-Network, Multi-Features Study

Open access
Darabant, Adrian Sergiu, Borza, Diana, Danescu, Radu · 2021 · Mathematics

The DOI matches its Crossref record; 57 of the 57 references it cites were found in a registry or an index.

Abstract

The human face holds a privileged position in multi-disciplinary research as it conveys much information—demographical attributes (age, race, gender, ethnicity), social signals, emotion expression, and so forth. Studies have shown that due to the distribution of ethnicity/race in training datasets, biometric algorithms suffer from “cross race effect”—their performance is better on subjects closer to the “country of origin” of the algorithm. The contributions of this paper are two-fold: (a) first, we gathered, annotated and made public a large-scale database of (over 175,000) facial images by automatically crawling the Internet for celebrities’ images belonging to various ethnicity/races, and (b) we trained and compared four state of the art convolutional neural networks on the problem of race and ethnicity classification. To the best of our knowledge, this is the largest, data-balanced, publicly-available face database annotated with race and ethnicity information. We also studied the impact of various face traits and image characteristics on the race/ethnicity deep learning classification methods and compared the obtained results with the ones extracted from psychological studies and anthropomorphic studies. Extensive tests were performed in order to determine the facial features to which the networks are sensitive to. These tests and a recognition rate of 96.64% on the problem of human race classification demonstrate the effectiveness of the proposed solution.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3390/math9020195

Cites

57 of 57 references verified0 citation errors · 0 not checkable

Statistics

Figures

Image 1
CC-BY · From the PDF
(b) Asian subjects
(b) Asian subjectsCC-BY · From the PDF
(c) Caucasian subjects
(c) Caucasian subjectsCC-BY · From the PDF
Figure 2 Distribution of the annotated labels in the gathered dataset.
Figure 2 Distribution of the annotated labels in the gathered dataset.CC-BY · From the PDF
Image 5
CC-BY · From the PDF
Image 6
CC-BY · From the PDF
Image 7
CC-BY · From the PDF
Afro-American as Caucasian
Afro-American as CaucasianCC-BY · From the PDF
Afro-American as Indian
Afro-American as IndianCC-BY · From the PDF
Image 10
CC-BY · From the PDF
Image 11
CC-BY · From the PDF
Image 12
CC-BY · From the PDF
Image 13
CC-BY · From the PDF
Image 14
CC-BY · From the PDF
Image 15
CC-BY · From the PDF
Image 16
CC-BY · From the PDF
Caucasian as Caucasian
Caucasian as CaucasianCC-BY · From the PDF
Image 18
CC-BY · From the PDF
Caucasian as Indian
Caucasian as IndianCC-BY · From the PDF
Image 20
CC-BY · From the PDF
Image 21
CC-BY · From the PDF
Indian as Caucasian
Indian as CaucasianCC-BY · From the PDF
Indian as Indian
Indian as IndianCC-BY · From the PDF
Figure 5 Masking procedure and heatmaps obtained after iteratively masking the images. ( a ). Masking procedure: the images is iteratively masked with a gray rectangle and the probability of the correct class is used to generate a heat-map ( b ). Heat-map for the Asian class. ( c ). Heat-map for the African-American class. ( d ). Heat-map for the Caucasian class. ( e ) Heat-map for the Indian class.
Figure 5 Masking procedure and heatmaps obtained after iteratively masking the images. ( a ). Masking procedure: the images is iteratively masked with a gray rectangle and the probability of the correct class is used to generate a heat-map ( b ). Heat-map for the Asian class. ( c ). Heat-map for the African-American class. ( d ). Heat-map for the Caucasian class. ( e ) Heat-map for the Indian class.CC-BY · From the PDF
Image 25
CC-BY · From the PDF
Figure 6 Radial blur: the input region of interest ( a ) is combined with the blurred region of interest ( c ), based on the weights defined by the blur mask ( b ). The result is illustrated in subfigure ( d ).
Figure 6 Radial blur: the input region of interest ( a ) is combined with the blurred region of interest ( c ), based on the weights defined by the blur mask ( b ). The result is illustrated in subfigure ( d ).CC-BY · From the PDF
Image 27
CC-BY · From the PDF
Image 28
CC-BY · From the PDF
Figure 7 Image alterations. ( a ) The 68 facial landmarks used in the localization of facial regions. ( b ) occlusion of the eyes. ( c ) radial blur or the eye area. ( d ) occlusion of the nostril area. ( e ) radial blur or the nostril area. ( f ) occlusion of the mouth area. ( g ) radial blur of the mouth area. ( h ) Gaussian blur of the entire image. ( i ) Increase brightness with a factor of 1.5. ( j ) Decrease brightness with a factor of 0.5.
Figure 7 Image alterations. ( a ) The 68 facial landmarks used in the localization of facial regions. ( b ) occlusion of the eyes. ( c ) radial blur or the eye area. ( d ) occlusion of the nostril area. ( e ) radial blur or the nostril area. ( f ) occlusion of the mouth area. ( g ) radial blur of the mouth area. ( h ) Gaussian blur of the entire image. ( i ) Increase brightness with a factor of 1.5. ( j ) Decrease brightness with a factor of 0.5.CC-BY · From the PDF
Figure 8 Some examples of Near Infrared (NIR) images from the LHDF dataset: ( a ) indoor, 1 m camera distance. ( b ). outdoor, 60 m camera distance, ( c ). outdoor, 100 m camera distance, ( d ). outdoor, 150 m camera distance.
Figure 8 Some examples of Near Infrared (NIR) images from the LHDF dataset: ( a ) indoor, 1 m camera distance. ( b ). outdoor, 60 m camera distance, ( c ). outdoor, 100 m camera distance, ( d ). outdoor, 150 m camera distance.CC-BY · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 57 references

Research works printed but not linked: 2

  1. Tin, H.H.K., and Sein, M.M. (2013, January 21–23). Automatic Race Identification from Face Images in Myanmar. Proceedings of the The First International Conference on Energy Environment and Human Engineering (ICEEHE 2013), Yangon, Myanmar.
  2. Howard, A., Sandler, M., Chu, G., Chen, L.C., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., and Vasudevan, V. (November, January 27). Searching for mobilenetv3. Proceedings of the IEEE International Conference on Computer Vision, Seoul, Korea.

Cited by

See all 26

Metadata

Authors
Darabant, A. S., Borza, D., & Danescu, R.
Affiliations
  • Computer Science Department, Babes Bolyai University, 1st Mihail Kogălniceanu Street, 400084 Cluj-Napoca, Romania
  • Babeș-Bolyai University (RO)
  • Technical University of Cluj-Napoca (RO)
Corresponding author
Borza, Diana
Venue
Mathematics
Publisher
MDPI AG
Published
2021-01-19
Volume
9
Issue
2
Pages
195
ISSN
2227-7390 (ISSN-L)
OpenAlex
W3122359801
Cited by
24
References
49
Citations by year
2026: 2 · 2025: 4 · 2024: 5 · 2023: 6 · 2022: 4 · 2021: 3
FWCI
1.62
Type
journal article
Language
en
Open access
Locations
  • Mathematics (publishedVersion, cc-by)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face and Expression Recognition
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Face Recognition and Perception
Keywords
racial classification, ethnic classification, face images, cross-race effect, convolutional neural networks, facial image database, celebrity image, image annotation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Recognizing Human Races through Machine Learning—A Multi-Network, Multi-Features Study. https://app.cochat.ai/papers/10.3390/math9020195
The paper
Darabant, A. S., Borza, D., & Danescu, R. (2021). Recognizing Human Races through Machine Learning—A Multi-Network, Multi-Features Study. Mathematics. https://doi.org/10.3390/math9020195

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.