POLO -- Point-based, multi-class animal detection

TL;DR

POLO is a novel multi-class animal detection model trained solely on point labels, reducing annotation effort while improving counting accuracy in drone imagery compared to standard methods.

Contribution

POLO introduces a point-label training approach for multi-class animal detection, modifying YOLOv8 for better accuracy with less annotation effort.

Findings

POLO outperforms YOLOv8 in animal counting accuracy.

POLO reduces annotation costs for wildlife surveys.

POLO effectively detects thousands of waterfowl in drone images.

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.