Classifying geospatial objects from multiview aerial imagery using semantic meshes

arXiv:2405.09544v1 Announce Type: new
Abstract: Aerial imagery is increasingly used in Earth science and natural resource management as a complement to labor-intensive ground-based surveys. Aerial systems can collect overlapping images that provide multiple views of each location from different perspectives. However, most prediction approaches (e.g. for tree species classification) use a single, synthesized top-down "orthomosaic" image as input that contains little to no information about the vertical aspects of objects and may include processing artifacts. We propose an alternate approach …

abstract aerial arxiv classification cs.cv earth geospatial however images labor location management meshes multiple natural objects perspectives prediction resource management science semantic species surveys systems tree type