HIGHLIGHTS
- who: With Georeferenced Images and collaborators from the (UNIVERSITY) have published the Article: Guiding Labelling Effort for Efficient Learning With Georeferenced Images, in the Journal: (JOURNAL)
- what: The authors assess the method`s sensitivities to design options using four different ground-truthed datasets of environmental monitoring where these include various scenes in aerial and seafloor imagery. The main reasons for this are the high sensitivity of image appearance to environmental conditions (e_g, lighting, atmosphere/water turbidity), observation variables (e_g, range to target, spatial resolution, observation footprint), the large variability in the appearance of unstructured scenes . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.