Gesture Matters: Pedestrian Gesture Recognition for AVs Through Skeleton Pose Evaluation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908822555066368 |
|---|---|
| author | Mahdi, Alif Rizqullah Rezaei, Mahdi Merat, Natasha |
| author_facet | Mahdi, Alif Rizqullah Rezaei, Mahdi Merat, Natasha |
| contents | Gestures are a key component of non-verbal communication in traffic, often helping pedestrian-to-driver interactions when formal traffic rules may be insufficient. This problem becomes more apparent when autonomous vehicles (AVs) struggle to interpret such gestures. In this study, we present a gesture classification framework using 2D pose estimation applied to real-world video sequences from the WIVW dataset. We categorise gestures into four primary classes (Stop, Go, Thank & Greet, and No Gesture) and extract 76 static and dynamic features from normalised keypoints. Our analysis demonstrates that hand position and movement velocity are especially discriminative in distinguishing between gesture classes, achieving a classification accuracy score of 87%. These findings not only improve the perceptual capabilities of AV systems but also contribute to the broader understanding of pedestrian behaviour in traffic contexts. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_08479 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Gesture Matters: Pedestrian Gesture Recognition for AVs Through Skeleton Pose Evaluation Mahdi, Alif Rizqullah Rezaei, Mahdi Merat, Natasha Computer Vision and Pattern Recognition Artificial Intelligence Emerging Technologies Human-Computer Interaction Machine Learning Gestures are a key component of non-verbal communication in traffic, often helping pedestrian-to-driver interactions when formal traffic rules may be insufficient. This problem becomes more apparent when autonomous vehicles (AVs) struggle to interpret such gestures. In this study, we present a gesture classification framework using 2D pose estimation applied to real-world video sequences from the WIVW dataset. We categorise gestures into four primary classes (Stop, Go, Thank & Greet, and No Gesture) and extract 76 static and dynamic features from normalised keypoints. Our analysis demonstrates that hand position and movement velocity are especially discriminative in distinguishing between gesture classes, achieving a classification accuracy score of 87%. These findings not only improve the perceptual capabilities of AV systems but also contribute to the broader understanding of pedestrian behaviour in traffic contexts. |
| title | Gesture Matters: Pedestrian Gesture Recognition for AVs Through Skeleton Pose Evaluation |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Emerging Technologies Human-Computer Interaction Machine Learning |
| url | https://arxiv.org/abs/2602.08479 |