TFW: Annotated Thermal Faces in the Wild Dataset

Askat Kuzdeuov, Dana Aubakirova, Darina Koishigarina, Huseyin Atakan Varol

Research output: Contribution to journalArticlepeer-review

9 Citations (Scopus)


Face detection and subsequent localization of facial landmarks are the primary steps in many face applications. Numerous algorithms and benchmark datasets have been introduced to develop robust models for the visible domain. However, varying conditions of illumination still pose challenging problems. In this regard, thermal cameras are employed to address this problem, because they operate on longer wavelengths. However, thermal face and facial landmark detection in the wild is an open research problem because most of the existing thermal datasets were collected in controlled environments. In addition, many of them were not annotated with face bounding boxes and facial landmarks. In this work, we present a thermal face dataset with manually labeled bounding boxes and facial landmarks to address these problems. The dataset contains 9,982 images of 147 subjects collected under controlled and uncontrolled conditions. As a baseline, we trained the YOLOv5 (Jocher, 2020) object detection model and its adaptation for face detection, YOLO5Face (Qi et al., 2021), on our dataset. In addition to our test set, we evaluated the models on the external RWTH-Aachen (Kopaczka et al., 2019) thermal face dataset to show the efficacy of our dataset. We have made the dataset, source code, and pre-trained models publicly available at to bolster research in thermal face analysis.

Original languageEnglish
Pages (from-to)2084-2094
Number of pages11
JournalIEEE Transactions on Information Forensics and Security
Publication statusPublished - 2022


  • Labeled thermal faces in the wild
  • YOLOv5
  • thermal face detection
  • thermal facial landmark detection

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality
  • Computer Networks and Communications


Dive into the research topics of 'TFW: Annotated Thermal Faces in the Wild Dataset'. Together they form a unique fingerprint.

Cite this