We are happy to announce that our research paper entitled “OpenThermalPose2: Extending the Open-Source Annotated Thermal Human Pose Dataset With More Data, Subjects, and Poses” (Askat Kuzdeuov, Miras Zakaryanov, Alim Tleuliyev, and Prof. Huseyin Atakan Varol) has been accepted for publication in the IEEE Transactions on Biometrics, Behavior, and Identity Science (Q1). 

Human pose estimation has applications in action recognition, human-robot interaction, motion capture, augmented reality, sports analytics, and healthcare. Numerous datasets and deep learning models have been developed for human pose estimation within the visible domain. However, poor lighting conditions and privacy issues persist. These challenges can be addressed using thermal cameras; however, there is a limited number of annotated thermal human pose datasets for training deep learning models.

To address this limitation, we extended our OpenThermalPose dataset by incorporating more thermal images, human instances, and poses. The extended dataset, OpenThermalPose2, contains 11,391 thermal images and 21,125 human instances annotated with bounding boxes and 17 anatomical keypoints. We trained the YOLOv8-pose and YOLO11-pose models to demonstrate the dataset’s effectiveness. Additionally, we optimized these models and deployed them on an NVIDIA Jetson AGX Orin 64GB for real-world applications. The dataset and pre-trained models are publicly available to support further research in this domain.

To read the full paper please visit IEEE Xplore. The dataset and pre-trained models are available at GitHub.