Simultaneous segmentation and classification of human actions in video streams using deeply optimized Hough transform

Adrien Chan-Hon-Tong (1) , Catherine Achard (2) , Laurent Lucat (1)

1 DIASI (CEA, LIST) - Département Intelligence Ambiante et Systèmes Interactifs
2 ISIR - Institut des Systèmes Intelligents et de Robotique

Most researches on human activity recognition do not take into account the temporal localization of actions. In this paper, a new method is designed to model both actions and their temporal domains. This method is based on a new Hough method which outperforms previous published ones on honeybee dataset thanks to a deeper optimization of the Hough variables. Experiments are performed to select skeleton features adapted to this method and relevant to capture human actions. With these features, our pipeline improves state-of-the-art performances on TUM dataset and outperforms baselines on several public datasets.

Mots clés: Temporal localization; Image recognition; Temporal domain; Hough; Video streaming; Image segmentation; Hough transforms; Classification (of information); Human actions; State-of-the-art performance; Human activity recognition

Domaines: Sciences de l'ingénieur [physics]

Adrien Chan-Hon-Tong, Catherine Achard, Laurent Lucat. Simultaneous segmentation and classification of human actions in video streams using deeply optimized Hough transform. Pattern Recognition, 2014, 47, pp.3807-3818. ⟨10.1016/j.patcog.2014.05.010⟩. ⟨cea-01818435⟩