Year: 2023
Type: book-chapter
Abstract: Visitor tracking has become a de facto standard for evaluating the success of exhibitions in museums (Yalowitz/Bronnenkraut 2009, 47).The data collection required for this analysis is, however, usually very labour-intensive or requires a costly setup (ibid.).But such systems usually do not gather, in particular, information about the visitor, like gender, age, and apparent interests.This information is instead painstakingly collected by means of questionnaires (ibid.).We propose a simple and costefficient automatic visitor monitoring pipeline to capture not only visitor trajectories, similar to established products, but also personal parameters generally only collected through questionnaires.Furthermore, we evaluate parts of this pipeline using state-of-the-art methods. Related WorkVarious methods can be used to conduct visitor tracking.The paper-and-pencil method is historically the most common (Yalowitz/Bronnenkraut 2009, 52-53) and requires a researcher to collect visitor data by hand.Modern methods feature radio technologies like Bluetooth, ultra-wide band, or wireless local area networks (De Angelis and Santoni 2022, 8).Older designs, in contrast, rely on infrared light and optical tracking approaches (Kuflik/Lanir/Dim et al. 2011, 375-76).But with recent advances in deep learning and person detection in images and videos, especially in the context of museum research (Bartoli et al. 2015, 19-27), visitor tracking by means of optical methods seems to have become a feasible alternative for visitor studies.
Source: transcript Verlag eBooks
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Funder: Deutsche Forschungsgemeinschaft
Awards: G353111149, G8663067234
OpenAlex ID: W4389938567