TY - GEN
T1 - SF-TL54
T2 - 2022 IEEE/SICE International Symposium on System Integration, SII 2022
AU - Kuzdeuov, Askat
AU - Koishigarina, Darina
AU - Aubakirova, Dana
AU - Abushakimova, Saniya
AU - Varol, Huseyin Atakan
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Facial landmark detection is a cornerstone in many facial analysis tasks such as face recognition, drowsiness detection, and facial expression recognition. Numerous methodologies were introduced to achieve accurate and efficient facial landmark localization in visual images. However, there are only several works that address facial landmark detection in thermal images. The main challenge is the limited number of annotated datasets. In this work, we present a thermal face dataset with annotated face bounding boxes and facial landmarks. The dataset contains 2, 556 thermal images of 142 individuals, where each thermal image is paired with the corresponding visual image. To the best of our knowledge, our dataset is the largest in terms of the number of individuals. In addition, our dataset can be employed for tasks such as thermal-to-visual image translation, thermal-visual face recognition, and others. We trained two models for the facial landmark detection task to show the efficacy of our dataset. The first model is a classic machine learning model based on an ensemble of regression trees. The second model is a deep learning model based on the U-net architecture. The dataset, annotations, source code, and pre-trained models are publicly available to advance research in thermal face analysis.
AB - Facial landmark detection is a cornerstone in many facial analysis tasks such as face recognition, drowsiness detection, and facial expression recognition. Numerous methodologies were introduced to achieve accurate and efficient facial landmark localization in visual images. However, there are only several works that address facial landmark detection in thermal images. The main challenge is the limited number of annotated datasets. In this work, we present a thermal face dataset with annotated face bounding boxes and facial landmarks. The dataset contains 2, 556 thermal images of 142 individuals, where each thermal image is paired with the corresponding visual image. To the best of our knowledge, our dataset is the largest in terms of the number of individuals. In addition, our dataset can be employed for tasks such as thermal-to-visual image translation, thermal-visual face recognition, and others. We trained two models for the facial landmark detection task to show the efficacy of our dataset. The first model is a classic machine learning model based on an ensemble of regression trees. The second model is a deep learning model based on the U-net architecture. The dataset, annotations, source code, and pre-trained models are publicly available to advance research in thermal face analysis.
UR - http://www.scopus.com/inward/record.url?scp=85126183664&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85126183664&partnerID=8YFLogxK
U2 - 10.1109/SII52469.2022.9708901
DO - 10.1109/SII52469.2022.9708901
M3 - Conference contribution
AN - SCOPUS:85126183664
T3 - 2022 IEEE/SICE International Symposium on System Integration, SII 2022
SP - 748
EP - 753
BT - 2022 IEEE/SICE International Symposium on System Integration, SII 2022
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 9 January 2022 through 12 January 2022
ER -