Corpus ID: 12447456

An Exploratory Study of Human Performance in Image Geolocation Tasks

  title={An Exploratory Study of Human Performance in Image Geolocation Tasks},
  author={Sneha Mehta and Chris North and Kurt Luther},
Identifying the precise location where a photo was taken is an important task in domains ranging from journalism to counter-terrorism. Yet, geolocation of arbitrary images is difficult for computer vision techniques and time-consuming for expert analysts. Understanding how humans perform geolocation can suggest rich opportunities for improvement, but little is known about their processes. This paper presents an exploratory study of image geolocation tasks performed by novice and expert humans… CONTINUE READING

Figures from this paper.


Publications referenced by this paper.
IM2GPS: estimating geographic information from a single image
  • 846
  • Open Access
PlaNet - Photo Geolocation with Convolutional Neural Networks
  • 242
  • Open Access
Learning deep representations for ground-to-aerial geolocalization
  • 203
  • Open Access
Human Computation in Citizen Science
  • 17
VizWiz: nearly real-time answers to visual questions
  • 310
Platemate: crowdsourcing nutritional analysis from food photographs
  • 237
  • Open Access
Geolocating Tunisian Jihadists in Raqqa
  • 2014
Human Factors of Wayfinding in Navigation
  • 36