May, Giacomo; Dalsasso, Emanuele; Kellenberger, Benjamin; Tuia, Devis; (2024) POLO - Point-based, multi-class animal detection. In: Del Bue, A, A and Canton, C, C and Pont-Tuset, J and Tommasi, T, (eds.) Computer Vision – ECCV 2024 Workshops. (pp. pp. 169-177). Springer, Cham: Cham, Switzerland. (In press).

Text 2410.11741v1.pdf - Accepted Version Access restricted to UCL open access staff until 13 May 2026. Download (2MB)

Abstract

Automated wildlife surveys based on drone imagery and object detection technology are a powerful and increasingly popular tool in conservation biology. Most detectors require training images with annotated bounding boxes, which are tedious, expensive, and not always unambiguous to create. To reduce the annotation load associated with this practice, we develop POLO, a multi-class object detection model that can be trained entirely on point labels. POLO is based on simple, yet effective modifications to the YOLOv8 architecture, including alterations to the prediction process, training losses, and post-processing. We test POLO on drone recordings of waterfowl containing up to multiple thousands of individual birds in one image and compare it to a regular YOLOv8. Our experiments show that at the same annotation cost, POLO achieves improved accuracy in counting animals in aerial imagery.

Download activity - last month

Export as XMLJSONCSV

Download activity - last 12 months

Export as CSVJSONXML

Downloads by country - last 12 months

Export as CSVJSONXML

Archive Staff Only

View Item