Skip to main content
placeholder image

Large-scale Continuous Gesture Recognition Using Convolutional Neural Networks

Conference Paper


Abstract


  • This paper addresses the problem of continuous gesture recognition from sequences of depth maps using Convolutional Neural networks (ConvNets). The proposed method first segments individual gestures from a depth sequence based on quantity of movement (QOM). For each segmented gesture, an Improved Depth Motion Map (IDMM), which converts the depth sequence into one image, is constructed and fed to a ConvNet for recognition. The IDMM effectively encodes both spatial and temporal information and allows the fine-tuning with existing ConvNet models for classification without introducing millions of parameters to learn. The proposed method is evaluated on the Large-scale Continuous Gesture Recognition of the ChaLearn Looking at People (LAP) challenge 2016. It achieved the performance of 0.2655 (Mean Jaccard Index) and ranked 3rd place in this challenge.

Authors


  •   Wang, Pichao (external author)
  •   Li, Wanqing
  •   Liu, Song (external author)
  •   Zhang, Yuyao (external author)
  •   Gao, Zhimin (external author)
  •   Ogunbona, Philip O.

Publication Date


  • 2016

Citation


  • Wang, P., Li, W., Liu, S., Zhang, Y., Gao, Z. & Ogunbona, P. (2016). Large-scale Continuous Gesture Recognition Using Convolutional Neural Networks. Proceedings - 23rd International Conference on Pattern Recognition (ICPR) (pp. 13-18). United States: IEEE.

Scopus Eid


  • 2-s2.0-85019114823

Start Page


  • 13

End Page


  • 18

Abstract


  • This paper addresses the problem of continuous gesture recognition from sequences of depth maps using Convolutional Neural networks (ConvNets). The proposed method first segments individual gestures from a depth sequence based on quantity of movement (QOM). For each segmented gesture, an Improved Depth Motion Map (IDMM), which converts the depth sequence into one image, is constructed and fed to a ConvNet for recognition. The IDMM effectively encodes both spatial and temporal information and allows the fine-tuning with existing ConvNet models for classification without introducing millions of parameters to learn. The proposed method is evaluated on the Large-scale Continuous Gesture Recognition of the ChaLearn Looking at People (LAP) challenge 2016. It achieved the performance of 0.2655 (Mean Jaccard Index) and ranked 3rd place in this challenge.

Authors


  •   Wang, Pichao (external author)
  •   Li, Wanqing
  •   Liu, Song (external author)
  •   Zhang, Yuyao (external author)
  •   Gao, Zhimin (external author)
  •   Ogunbona, Philip O.

Publication Date


  • 2016

Citation


  • Wang, P., Li, W., Liu, S., Zhang, Y., Gao, Z. & Ogunbona, P. (2016). Large-scale Continuous Gesture Recognition Using Convolutional Neural Networks. Proceedings - 23rd International Conference on Pattern Recognition (ICPR) (pp. 13-18). United States: IEEE.

Scopus Eid


  • 2-s2.0-85019114823

Start Page


  • 13

End Page


  • 18