OAK-D Synthethic Pose Dataset

<p>"OAK-D Synthethic Pose Dataset" is a synthetic image dataset of humans, annotated with 2d and 3d keypoints for a OAK-D like camera configuration.</p> <p>The dataset is part of the "3D Stereo Body Pose Estimation" project, originated in the context of the activities of the postgraduate course IA904 - Model Project in Visual Computing, offered in the first semester of 2024, at Unicamp, under the supervision of Prof. Dr. Leticia Rittner and Prof. Paula D. Paro Costa, both from the Department of Computer and Automation Engineering (DCA) of the Faculty of Electrical and Computer Engineering (FEEC).</p> <p>The <a href="https://github.com/Disciplinas-FEEC/IA904-2024S1/tree/main/projetos/3DStereoBodyPoseEstimation">project page</a> have a full description of the project (portuguese).</p>