Agritrop
Accueil

Overview of BirdCLEF 2024: Acoustic identification of under-studied bird species in the Western Ghats

Kahl Stefan, Denton Tom, Klinck Holger, Ramesh Vijay, Joshi Viral, Srivathsa Meghana, Anand Akshay, Arvind Chiti, Cp Harikrishnan, Sawant Suyash, Vv Robin, Glotin Hervé, Goeau Hervé, Vellinga Willem-Pier, Planqué Robert, Joly Alexis. 2024. Overview of BirdCLEF 2024: Acoustic identification of under-studied bird species in the Western Ghats. In : Working Notes of the Conference and Labs of the Evaluation Forum (CLEF 2024). Faggioli Guglielmo (ed.), Ferro Nicola (ed.), Galuščáková Petra (ed.), García Seco de Herrera Alba (ed.). Aachen : CEUR-WS, 1948-1957. Conference and Labs of the Evaluation Forum (CLEF 2024), Grenoble, France, 9 Septembre 2024/12 Septembre 2024.

Communication avec actes
[img]
Prévisualisation
Version publiée - Anglais
Utilisation soumise à autorisation de l'auteur ou du Cirad.
613024.pdf

Télécharger (29MB) | Prévisualisation

Résumé : The BirdCLEF 2024 challenge focused on the acoustic identification of understudied bird species in the Western Ghats, a biodiversity hotspot in India. This edition aimed to advance passive acoustic monitoring by tasking participants with developing reliable systems for detecting and identifying bird vocalizations from extensive soundscape recordings. Using training data provided by the Xeno-Canto community and new unlabeled soundscapes from the Western Ghats, participants addressed the challenges of domain adaptation and limited training data for many species. Participants employed techniques such as pseudo-labeling, test-time augmentation, and diverse ensembles, significantly improving model performance. Notable strategies also included the use of single-class cross-entropy and Contrastive Adversarial Domain (CAD) bottlenecks, which provided innovative solutions to acoustic data analysis challenges. The highest-scoring submission achieved an ROC-AUC score of 0.690 on the private leaderboard (0.738 on the public leaderboard), with the top 10 systems differing by only 1.5% in their scores.

Mots-clés libres : LifeCLEF, Bird, Song, Call, Species, Retrieval, Audio, Collection, Identification, Fine-grained classification, Evalution, Benchmark, Bioascoustics, Passive Acoustic Monitoring, PAM

Auteurs et affiliations

  • Kahl Stefan, Cornell University (USA)
  • Denton Tom, Google Deepmind (USA)
  • Klinck Holger, Cornell University (USA)
  • Ramesh Vijay, Cornell University (USA)
  • Joshi Viral, IISER (IND)
  • Srivathsa Meghana, Project Dhvani (IND)
  • Anand Akshay, Project Dhvani (IND)
  • Arvind Chiti, IISER (IND)
  • Cp Harikrishnan, IISER (IND)
  • Sawant Suyash, IISER (IND)
  • Vv Robin, IISER (IND)
  • Glotin Hervé, Université de Toulon et du Var (FRA)
  • Goeau Hervé, CIRAD-BIOS-UMR AMAP (FRA)
  • Vellinga Willem-Pier, Xeno-canto foundation (NLD)
  • Planqué Robert, Xeno-canto foundation (NLD)
  • Joly Alexis, INRIA (FRA)

Source : Cirad-Agritrop (https://agritrop.cirad.fr/613024/)

Voir la notice (accès réservé à Agritrop) Voir la notice (accès réservé à Agritrop)

[ Page générée et mise en cache le 2025-09-22 ]