MuseumVisitors: A Dataset for Pedestrian and Group Detection, Gaze Estimation and Behavior Understanding

Abstract

In this paper we describe a new dataset, under construction,
\nacquired inside the National Museum of Bargello in
\nFlorence. It was recorded with three IP cameras at a resolution
\nof 1280 × 800 pixels and an average framerate of
\nfive frames per second. Sequences were recorded following
\ntwo scenarios. The first scenario consists of visitors watching
\ndifferent artworks (individuals), while the second one
\nconsists of groups of visitors watching the same artworks
\n(groups).
\nThis dataset is specifically designed to support research
\non group detection, occlusion handling, tracking,
\nre-identification and behavior analysis. In order to ease
\nthe annotation process we designed a user friendly web interface
\nthat allows to annotate: bounding boxes, occlusion
\narea, body orientation and head gaze, group belonging, and
\nartwork under observation. We provide a comparison with
\nother existing datasets that have group and occlusion annotations.
\nIn order to assess the difficulties of this dataset we
\nhave also performed some tests exploiting seven representative
\nstate-of-the-art pedestrian detectors.

Cite

Text

Bartoli et al. "MuseumVisitors: A Dataset for Pedestrian and Group Detection, Gaze Estimation and Behavior Understanding." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2015. doi:10.1109/CVPRW.2015.7301279

Markdown

[Bartoli et al. "MuseumVisitors: A Dataset for Pedestrian and Group Detection, Gaze Estimation and Behavior Understanding." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2015.](https://mlanthology.org/cvprw/2015/bartoli2015cvprw-museumvisitors/) doi:10.1109/CVPRW.2015.7301279

BibTeX

@inproceedings{bartoli2015cvprw-museumvisitors,
  title     = {{MuseumVisitors: A Dataset for Pedestrian and Group Detection, Gaze Estimation and Behavior Understanding}},
  author    = {Bartoli, Federico and Lisanti, Giuseppe and Seidenari, Lorenzo and Karaman, Svebor and Del Bimbo, Alberto},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2015},
  pages     = {19-27},
  doi       = {10.1109/CVPRW.2015.7301279},
  url       = {https://mlanthology.org/cvprw/2015/bartoli2015cvprw-museumvisitors/}
}