ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume IV-2/W5
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., IV-2/W5, 341–348, 2019
https://doi.org/10.5194/isprs-annals-IV-2-W5-341-2019
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., IV-2/W5, 341–348, 2019
https://doi.org/10.5194/isprs-annals-IV-2-W5-341-2019

  29 May 2019

29 May 2019

NON-RIGID MULTI-BODY TRACKING IN RGBD STREAMS

K. X. Dai1, H. Guo1, P. Mordohai2, F. Marinello3, A. Pezzuolo3, Q. L. Feng1, and Q. D. Niu1 K. X. Dai et al.
  • 1College of Land Science and Technology, China Agricultural University, Beijing 100083, China
  • 2Department of Computer Science, Stevens Institute of Technology, NJ 07030, USA
  • 3Department of Land, Environment, Agriculture and Forestry, University of Padova, Legnaro(PD) 35020, Italy

Keywords: Non-rigid, Multi-body tracking, RGBD, Point clouds

Abstract. To efficiently collect training data for an off-the-shelf object detector, we consider the problem of segmenting and tracking non-rigid objects from RGBD sequences by introducing the spatio-temporal matrix with very few assumptions – no prior object model and no stationary sensor. Spatial temporal matrix is able to encode not only spatial associations between multiple objects, but also component-level spatio temporal associations that allow the correction of falsely segmented objects in the presence of various types of interaction among multiple objects. Extensive experiments over complex human/animal body motions with occlusions and body part motions demonstrate that our approach substantially improves tracking robustness and segmentation accuracy.