Continuous detection of human fall using multimodal features from Kinect sensors in scalable environment.

Journal: Computer methods and programs in biomedicine
Published Date:

Abstract

BACKGROUND AND OBJECTIVES: Automatic detection of human fall is a key problem in video surveillance and home monitoring. Existing methods using unimodal data (RGB / depth / skeleton) may suffer from the drawbacks of inadequate lighting condition or unreliability. Besides, most of proposed methods are constrained to a small space with off-line video stream.

Authors

  • Thanh-Hai Tran
    International Research Institute MICA, HUST-CNRS/UMI-2954-GRENOBLE INP, Hanoi University of Science and Technology, Hanoi, Vietnam. Electronic address: thanh-hai.tran@mica.edu.vn.
  • Thi-Lan Le
    International Research Institute MICA, HUST-CNRS/UMI-2954-GRENOBLE INP, Hanoi University of Science and Technology, Hanoi, Vietnam.
  • Van-Nam Hoang
    International Research Institute MICA, HUST-CNRS/UMI-2954-GRENOBLE INP, Hanoi University of Science and Technology, Hanoi, Vietnam.
  • Hai Vu
    International Research Institute MICA, HUST-CNRS/UMI-2954-GRENOBLE INP, Hanoi University of Science and Technology, Hanoi, Vietnam.