Biology subjects

Glotin, H.

Publications and source records attributed to Glotin, H..

3 recordsLinked to original sources

bioRxiv2025

Data-driven Sampling Strategies for Fine-Tuning Bird Detection Models

Passive Acoustic Monitoring 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, PAM collects vast amounts of data, and annotating bird vocalizations requires specialized expetise. 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 passive acoustic monitoring.

bioinformatics↗

bioRxiv2023

Deep audio embeddings for vocalisation clustering

The study of non-human animals communication systems generally relies on the transcription of vocal sequences using a finite set of discrete units. This set is referred to as a vocal repertoire, which is specific to a species or a sub-group of a species. When conducted by human experts, the formal description of vocal repertoires can be laborious and/or biased. This motivates computerised assistance for this procedure, for which machine learning algorithms represent a good opportunity. Unsupervised clustering algorithms are suited for grouping close points together, provided a relevant representation. This paper therefore studies a new method for encoding vocalisations, allowing for automatic clustering to alleviate vocal repertoire characterisation. Borrowing from deep representation learning, we use a convolutional auto-encoder network to learn an abstract representation of vocalisations. We report on the quality of the learnt representation, as well as of state of the art methods, by quantifying their agreement with expert labelled vocalisation types from 7 datasets of other studies across 6 species (birds and marine mammals). With this benchmark, we demonstrate that using auto-encoders improves the relevance of vocalisation representation which serves repertoire characterisation using a very limited number of settings. We also publish a Python package for the bioacoustic community to train their own vocalisation auto-encoders or use a pretrained encoder to browse vocal repertoires and ease unit wise annotation.

bioinformatics↗

bioRxiv2021

Underwater photo-identification of marine megafauna: an identity card catalogue of sperm whales (Physeter macrocephalus) off Mauritius Island

The long-term monitoring of long-lived animal populations often requires individual identification. For cetacean populations, this identification is mostly based on morphological characters observable from a boat such as shape, spots and cuts of the back, caudal and dorsal fins. This is well suited for species easily displaying their caudal fins, such as the humpback whales Megaptera novaeangliae, or those whose skin pigmentation patterns enable individual identification. However, for elusive or shier species such as the sperm whales Physeter macrocephalus, this approach may be more challenging as individuals display a rather uniform skin pigmentation. They also do not show very often their caudal fin that must be photographed perpendicularly to the water surface, vertically and fully emerged, uneasing the individual identification from a boat. Immature sperm whales that usually have a caudal fin without any distinctive marks may sometimes be excluded from photo-identification catalogues. Within the framework of the Maubydick project, focusing on the long-term monitoring of sperm whales in Mauritius, passive underwater observation and video recording were used to identify long-lasting body markers (e.g., sex, ventral white markings, cut outs of fins) to improve individual identification. A catalogue of individual identity cards was developed and 38 individuals were recorded (six adult males, 18 adult females and 14 immatures). This catalogue was used in the field and enabled observers to record some nearly-daily and yearly recaptures. Advantages and disadvantages of this method are presented here. Such catalogues represent a robust baseline for conducting behavioural, genetic and acoustic studies in marine megafauna social species. Benefits of such newly acquired knowledge are of first importance to implement relevant conservation plans in the marine realm.

ecology↗