Synergies of Gait Recognition and Person Tracking

• Applicant: Professor Dr.-Ing. Gerhard Rigoll
• Subject Area: Image and Language Processing, Computer Graphics and Visualisation, Human Computer Interaction, Ubiquitous and Wearable Computing
• Term: from 2014 to 2019
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 262055636

• Final Report Year: 2018

Final Report Abstract

As one of the biometric features for human recognition, Gait has drawn significant attention in recent years, since it can be used to identify people from large distance. On the other hand, multiple people tracking aims to identify people while observed by one or several cameras. Both topics are challenging problems in video surveillance applications. In this project, we addressed the problem of collaborative gait recognition and people tracking since multipeople gait recognition and the multi-people tracking problems are closely related. We made efforts in these two research areas to be further developed and combined. The proposed combination of methods to identify a person on the basis of gait with modern tracking algorithms has the potential new safety systems produce, in particular can be applied for realistic scenarios where for example on Airports, there would be many people persecuted in the camera image.

Publications

• 2016, October. Pixel level tracking of multiple targets in crowded environments. In European Conference on Computer Vision (pp. 692-708). Springer, Cham
  Babaee, M., You, Y. and Rigoll, G.
  (See online at https://doi.org/10.1007/978-3-319-48881-3_49)
• 2016, September. Multi-view gait recognition using 3D convolutional neural networks. In Image Processing (ICIP), 2016 IEEE International Conference on (pp. 4165-4169)
  Wolf, T., Babaee, M. and Rigoll, G.
  (See online at https://doi.org/10.1109/ICIP.2016.7533144)
• 2017, October. View-Invariant Gait Representation Using Joint Bayesian Regularized Non-negative Matrix Factorization. In Computer Vision Workshop (ICCVW), 2017 IEEE International Conference on (pp. 2583-2589)
  Babaee, M. and Rigoll, G.
  
• 2017. Combined segmentation, reconstruction, and tracking of multiple targets in multi-view video sequences. Computer Vision and Image Understanding, 154, pp.166-181
  Babaee, M., You, Y. and Rigoll, G.
  (See online at https://doi.org/10.1016/j.cviu.2016.08.006)
• 2018. A deep convolutional neural network for video sequence background subtraction. Pattern Recognition, 76, pp.635-649
  Babaee, M., Dinh, D.T. and Rigoll, G.
  (See online at https://doi.org/10.1016/j.patcog.2017.09.040)