Head Pose Annotations Used in the LandPose-VLoc paper for 2D and 3D Face and Mouth Localization

This dataset contains the face and mouth landmark annotations used in the experiments reported in: "Landmark-Assisted Face and Mouth 3D Localization in Smart Spaces with a Monocular Camera" (submitted to Journal of Visual Communication and Image Representation). The files include: Estimated head pose in MATLAB (.mat) format for the LandFree-VLoc algorithm used in the paper, generated by the img2pose system (code at https://github.com/vitoralbiero/img2pose) Estimated facial landmarks in JSON format for the LandPose-VLoc algorithm used in the paper, generated with the system described in the paper "Multi-Task Head Pose Estimation in-the-Wild" by R. Valle, J. M. Buenaposada & L. Baumela (MT-HPE-W). Data were generated using the methodology described in the paper and were used for mouth position localization in 2D and 3D. The repository is intended to ensure full reproducibility of the reported experimental results.