Featureweighting in dynamic timewarping for gesture recognition in depth data

By Miguel Reyes, Gabriel Dominguez, Sergio Escalera. 2011

170 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Human-Computer Interaction, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Jun Liu
2. G. Wang
3. Guo-qing Guo
4. Ramalingam Chellappa
5. Luc Van Gool
6. Hossein Rahmani
7. Shuicheng Yan
8. Fei Han

In Human-Computer Interaction

1. Anuj Srivastava
2. Guo-qing Guo
3. Wenbin Chen
4. Mohamed Daoudi
5. Y. Zhu
6. Ramalingam Chellappa
7. Raviteja Vemulapalli
8. Hazem Wannous

In Signal Processing

1. Guo-qing Guo
2. Shuicheng Yan
3. William A. Hoff
4. Brian Reily
5. Fei Han
6. Feng Jiang
7. Y. Zhu
8. Wenbin Chen

Citations in the corpus

• Space-Time Representation of People Based on 3D Skeletal Data: A Review (2016, methodology): “[166] Vector of Joints Conc Lowlv Hand X X”
• Simultaneous Feature and Body-Part Learning for real-time robot awareness of human behaviors (2017, methodology): “Several methods were also proposed to select more descriptive human body joints [2], [10], [11], [20], [21], [22], [23].”
• Body Surface Context: A New Robust Feature for Action Recognition From Depth Videos (2014, methodology): “[29] adopted the relative coordinates of joints as a feature and Raptis et al.”
• Rolling Rotations for Recognizing Human Actions from 3D Skeletal Data (2016, background): “Joint-based approaches consider human skeleton as a set of points, and represent it using features like joint positions [7, 8, 12, 15, 20, 25], pairwise relative joint positions [33, 34, 35, 38], joint orientations in a fixed coordinate system [21, 23, 31, 36], etc.”
• Survey on 3D Hand Gesture Recognition (2016, methodology): “[48] presented a begin–end gesture recognition approach using feature weighting in the DTW framework.”
• Accurate 3D action recognition using learning on the Grassmann manifold (2015, methodology): “Reyes et al. [38] per-212 form DTW on a feature vector defined by 15 joints on a 3D human skeleton obtained using PrimeSense NiTE.”
• Fusing Multiple Features for Depth-Based Action Recognition (2015, methodology): “According to Ross and Jain [2003], feature-level fusion is usually conducted through feature normalization and feature selection or transformation because of the relationship between different feature sets and the curse of dimensionality [Donoho et al.”
• Multi-layered gesture recognition with Kinect (2015, methodology): “In the first stage, DTW is applied to align the reference and test sequences.”