Data-driven sampling strategies for fine-tuning bird detection modelsa).

PubMed, vol. 159, pp. 4891–4903

Abstract

Passive acoustic monitoring (PAM) has emerged as a promising tool for collecting ecological data, particularly in the context of bird population monitoring. Bird species can be automatically identified using pre-trained models, such as BirdNET. The performance of these models can be significantly improved through fine-tuning with annotated samples recorded in the specific acoustic conditions in which the microphones are deployed. However, passive acoustic monitoring (PAM) collects vast amounts of data, and annotating bird vocalizations requires specialized expertise. As a result, only a very small portion of the recordings can be effectively labeled. Selecting the most relevant samples to annotate in order to maximize performance in model fine-tuning remains a significant challenge. First, a regularization technique addresses the challenge of class imbalance during model fine-tuning. Next, a data-driven methodology is developed, introducing the influence score, which quantifies the impact of individual training samples on model performance to inform sampling strategies. A linear model is proposed to estimate the influence score for generalization to unseen data. Finally, several sampling strategies are compared, based on acoustic indices and predictions of the pre-trained model. Together, these contributions enable the identification of efficient annotation strategies to overcome the challenges of limited annotation resources in large-scale PAM.

Authors 6

1. Corentin Bernard

   Centre National de la Recherche Scientifique · Université de Toulon · Aix-Marseille Université · Laboratoire d’Informatique et Systèmes

   Affiliation as printed

   Univ Toulon, Aix Marseille Univ, CNRS, LIS, Toulon, France

   Univ Toulon, CIAN, Toulon, France

2. Ben McEwen

   Tilburg University

   Affiliation as printed

   Department of Intelligent Systems, Tilburg University, Tilburg, The Netherlands

3. Benjamin Cretois

   Norwegian Institute for Nature Research

   Affiliation as printed

   Norwegian Institute for Nature Research, Trondheim, Norway

4. Hervé Glotin

   Centre National de la Recherche Scientifique · Université de Toulon · Aix-Marseille Université · Laboratoire d’Informatique et Systèmes

   Affiliation as printed

   Univ Toulon, Aix Marseille Univ, CNRS, LIS, Toulon, France

   Univ Toulon, CIAN, Toulon, France

5. Dan Stowell Aachen Naturalis Biodiversity Center

   Naturalis Biodiversity Center · Tilburg University

   Affiliation as printed

   Department of Intelligent Systems, Tilburg University, Tilburg, The Netherlands

   Naturalis Biodiversity Center, Leiden, The Netherlands

6. Ricard Marxer

   Centre National de la Recherche Scientifique · Université de Toulon · Aix-Marseille Université · Laboratoire d’Informatique et Systèmes

   Affiliation as printed

   Univ Toulon, Aix Marseille Univ, CNRS, LIS, Toulon, France

   Univ Toulon, CIAN, Toulon, France

   CNRS, ILLS, Montreal, Canada