Real-time Tracking of Handheld Object from Color or Depth Images

1Rehabilitation Research Institute of Singapore

Choi and Christensen dataset

Qualitative results on synthetic sequence

Qualitative results on real sequence

RBOT dataset

Qualitative results on cat object for regular, dynamic light, noise and occlusion sequences

Qualitative results on cube object

Top without keypoint term: tracking fails frequently as shown by black out image, Bottom with keypoint term: has a more robust tracking with lesser black out image

DodecaPen dataset

Qualitative results on synthetic sequence

Qualitative results on using DodecaPen for handwriting task

BibTeX

@inproceedings{ObjTrack:2023,
      title     = {Real-time Tracking of Handheld Object from Color or Depth Images},
      author    = {Guan Ming, Lim and Prayook, Jatesiktat and Wei Tech, Ang},
      booktitle = {45th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)},
      year      = {2023}
    }