Skip to content
Back to outputs

Occlusion gesture recognition based on improved SSD

Research output: Contribution to journalArticlepeer-review

Standard

Occlusion gesture recognition based on improved SSD. / Liao, Shangchuan; Li, Gongfa; Wu, Hao; Jiang, Du; Liu, Ying ; Yun, Juntong; Liu, Yibo; Zhou, Dalin.

In: Concurrency and Computation: Practice and Experience, 19.10.2020.

Research output: Contribution to journalArticlepeer-review

Harvard

Liao, S, Li, G, Wu, H, Jiang, D, Liu, Y, Yun, J, Liu, Y & Zhou, D 2020, 'Occlusion gesture recognition based on improved SSD', Concurrency and Computation: Practice and Experience. https://doi.org/10.1002/cpe.6063

APA

Liao, S., Li, G., Wu, H., Jiang, D., Liu, Y., Yun, J., Liu, Y., & Zhou, D. (2020). Occlusion gesture recognition based on improved SSD. Concurrency and Computation: Practice and Experience. https://doi.org/10.1002/cpe.6063

Vancouver

Liao S, Li G, Wu H, Jiang D, Liu Y, Yun J et al. Occlusion gesture recognition based on improved SSD. Concurrency and Computation: Practice and Experience. 2020 Oct 19. https://doi.org/10.1002/cpe.6063

Author

Liao, Shangchuan ; Li, Gongfa ; Wu, Hao ; Jiang, Du ; Liu, Ying ; Yun, Juntong ; Liu, Yibo ; Zhou, Dalin. / Occlusion gesture recognition based on improved SSD. In: Concurrency and Computation: Practice and Experience. 2020.

Bibtex

@article{7cd377b8f5154251ab751cb0642c00f0,
title = "Occlusion gesture recognition based on improved SSD",
abstract = "Gesture recognition has always been a research hotspot in the field of human‐computer interaction. Its purpose is to realize the natural interaction with the machine by recognizing the semantics expressed by gesture. In the process of gesture recognition, the occlusion of gesture is an inevitable problem. In the process of gesture recognition, some or even all of the gesture features will be lost due to the occlusion of the gesture, resulting in the wrong recognition or even unrecognizability of the gesture. Therefore, it is of great significance to study gesture recognition under occlusion. The single shot multibox detector (SSD) algorithm is analyzed, and the front‐end network is compared. Mobilenets is selected as the front‐end network, and the Mobilenets‐SSD network is improved. In tensorflow environment, based on the improved network model, the self‐occlusion gesture and object occluding gesture are trained in color map, depth map, and color and depth fusion respectively. The recognition models of self‐occlusion gestures and object‐occlusion gestures in color map, depth map, and color and depth fusion are obtained. And compare and analyze the learning rate, loss function, and average accuracy of various models obtained for occlusion gesture recognition.",
keywords = "gesture recognition, human computer interaction, occlusion, SSD",
author = "Shangchuan Liao and Gongfa Li and Hao Wu and Du Jiang and Ying Liu and Juntong Yun and Yibo Liu and Dalin Zhou",
year = "2020",
month = oct,
day = "19",
doi = "10.1002/cpe.6063",
language = "English",
journal = "Concurrency and Computation: Practice and Experience",
issn = "1532-0626",
publisher = "John Wiley and Sons Ltd",

}

RIS

TY - JOUR

T1 - Occlusion gesture recognition based on improved SSD

AU - Liao, Shangchuan

AU - Li, Gongfa

AU - Wu, Hao

AU - Jiang, Du

AU - Liu, Ying

AU - Yun, Juntong

AU - Liu, Yibo

AU - Zhou, Dalin

PY - 2020/10/19

Y1 - 2020/10/19

N2 - Gesture recognition has always been a research hotspot in the field of human‐computer interaction. Its purpose is to realize the natural interaction with the machine by recognizing the semantics expressed by gesture. In the process of gesture recognition, the occlusion of gesture is an inevitable problem. In the process of gesture recognition, some or even all of the gesture features will be lost due to the occlusion of the gesture, resulting in the wrong recognition or even unrecognizability of the gesture. Therefore, it is of great significance to study gesture recognition under occlusion. The single shot multibox detector (SSD) algorithm is analyzed, and the front‐end network is compared. Mobilenets is selected as the front‐end network, and the Mobilenets‐SSD network is improved. In tensorflow environment, based on the improved network model, the self‐occlusion gesture and object occluding gesture are trained in color map, depth map, and color and depth fusion respectively. The recognition models of self‐occlusion gestures and object‐occlusion gestures in color map, depth map, and color and depth fusion are obtained. And compare and analyze the learning rate, loss function, and average accuracy of various models obtained for occlusion gesture recognition.

AB - Gesture recognition has always been a research hotspot in the field of human‐computer interaction. Its purpose is to realize the natural interaction with the machine by recognizing the semantics expressed by gesture. In the process of gesture recognition, the occlusion of gesture is an inevitable problem. In the process of gesture recognition, some or even all of the gesture features will be lost due to the occlusion of the gesture, resulting in the wrong recognition or even unrecognizability of the gesture. Therefore, it is of great significance to study gesture recognition under occlusion. The single shot multibox detector (SSD) algorithm is analyzed, and the front‐end network is compared. Mobilenets is selected as the front‐end network, and the Mobilenets‐SSD network is improved. In tensorflow environment, based on the improved network model, the self‐occlusion gesture and object occluding gesture are trained in color map, depth map, and color and depth fusion respectively. The recognition models of self‐occlusion gestures and object‐occlusion gestures in color map, depth map, and color and depth fusion are obtained. And compare and analyze the learning rate, loss function, and average accuracy of various models obtained for occlusion gesture recognition.

KW - gesture recognition

KW - human computer interaction

KW - occlusion

KW - SSD

U2 - 10.1002/cpe.6063

DO - 10.1002/cpe.6063

M3 - Article

JO - Concurrency and Computation: Practice and Experience

JF - Concurrency and Computation: Practice and Experience

SN - 1532-0626

ER -

ID: 22532623