{"ID":2863778,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.25042","arxiv_id":"2509.25042","title":"Fast Real-Time Pipeline for Robust Arm Gesture Recognition","abstract":"This paper presents a real-time pipeline for dynamic arm gesture recognition based on OpenPose keypoint estimation, keypoint normalization, and a recurrent neural network classifier. The 1 x 1 normalization scheme and two feature representations (coordinate- and angle-based) are presented for the pipeline. In addition, an efficient method to improve robustness against camera angle variations is also introduced by using artificially rotated training data. Experiments on a custom traffic-control gesture dataset demonstrate high accuracy across varying viewing angles and speeds. Finally, an approach to calculate the speed of the arm signal (if necessary) is also presented.","short_abstract":"This paper presents a real-time pipeline for dynamic arm gesture recognition based on OpenPose keypoint estimation, keypoint normalization, and a recurrent neural network classifier. The 1 x 1 normalization scheme and two feature representations (coordinate- and angle-based) are presented for the pipeline. In addition,...","url_abs":"https://arxiv.org/abs/2509.25042","url_pdf":"https://arxiv.org/pdf/2509.25042v1","authors":"[\"Milán Zsolt Bagladi\",\"László Gulyás\",\"Gergő Szalay\"]","published":"2025-09-29T16:57:56Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[]","has_code":false}