We present a novel method for monocular hand gesture recognition in ego-vision scenarios that deals with static and dynamic gestures and can achieve high accuracy results using a few positive samples. Specifically, we use and extend the dense trajectories approach that has been successfully introduced for action recognition. Dense features are extracted around regions selected by a new hand segmentation technique that integrates superpixel classification, temporal and spatial coherence. We extensively testour gesture recognition and segmentation algorithms on public datasets and propose a new dataset shot with a wearable camera. In addition, we demonstrate that our solution can work in near real-time on a wearable device.
We present a novel method for monocular hand gesture recognition in ego-vision scenarios that deals with static and dynamic gestures and can achieve high accuracy results using a few positive samples. Specifically, we use and extend the dense trajectories approach that has been successfully introduced for action recognition. Dense features are extracted around regions selected by a new hand segmentation technique that integrates superpixel classification, temporal and spatial coherence. We extensively testour gesture recognition and segmentation algorithms on public datasets and propose a new dataset shot with a wearable camera. In addition, we demonstrate that our solution can work in near real-time on a wearable device.
Gesture recognition in ego-centric videos using dense trajectories and hand segmentation / Baraldi, Lorenzo; Paci, Francesco; Serra, Giuseppe; Benini, Luca; Cucchiara, Rita. - STAMPA. - (2014), pp. 702-707. ((Intervento presentato al convegno 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2014 tenutosi a Columbus, Ohio nel 23-28 June 2014.
Titolo: | Gesture recognition in ego-centric videos using dense trajectories and hand segmentation |
Autori: | |
Data di pubblicazione: | 2014 |
Rivista: | |
Citazione: | Gesture recognition in ego-centric videos using dense trajectories and hand segmentation / Baraldi, Lorenzo; Paci, Francesco; Serra, Giuseppe; Benini, Luca; Cucchiara, Rita. - STAMPA. - (2014), pp. 702-707. ((Intervento presentato al convegno 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2014 tenutosi a Columbus, Ohio nel 23-28 June 2014. |
Abstract: | We present a novel method for monocular hand gesture recognition in ego-vision scenarios that deals with static and dynamic gestures and can achieve high accuracy results using a few positive samples. Specifically, we use and extend the dense trajectories approach that has been successfully introduced for action recognition. Dense features are extracted around regions selected by a new hand segmentation technique that integrates superpixel classification, temporal and spatial coherence. We extensively testour gesture recognition and segmentation algorithms on public datasets and propose a new dataset shot with a wearable camera. In addition, we demonstrate that our solution can work in near real-time on a wearable device. |
Handle: | http://hdl.handle.net/11390/1105589 |
ISBN: | 978-1-4799-4309-8 |
Appare nelle tipologie: | 4.1 Contributo in Atti di convegno |