Can I read Multispecies Animal Re-ID Using a Large Community-Curated Dataset on EtoBox?

Multispecies Animal Re-ID Using a Large Community-Curated Dataset by Otarashvili, Lasha; Subramanian, Tamilselvan; Holmberg, Jason; Levenson, J. J.; Stewart, Charles V. is a scholarly article available to read on EtoBox.

What is Multispecies Animal Re-ID Using a Large Community-Curated Dataset about?

Recent work has established the ecological importance of developing algorithms for identifying animals individually from images. Typically, a separate algorithm is trained for each species, a natural step but one that creates significant barriers to wide-spread use: (1) each effort is expensive, requiring data collection, data curation, and model training, deployment, and maintenance, (2) there is little training data for many species, and (3) commonalities in appearance across species are not exploited. We propose an alternative approach focused on training multi-species individual identification (re-id) models. We construct a dataset that includes 49 species, 37K individual animals, and 225K images, using this data to train a single embedding network for all species. Our model employs an EfficientNetV2 backbone and a sub-center ArcFace loss function with dynamic margins. We evaluate the performance of this multispecies model in several ways. Most notably, we demonstrate that it consistently outperforms models trained separately on each species, achieving an average gain of 12.5% in top-1 accuracy. Furthermore, the model demonstrates strong zero-shot performance and fine-tuning ca

Author

Otarashvili, Lasha; Subramanian, Tamilselvan; Holmberg, Jason; Levenson, J. J.; Stewart, Charles V.

Published

2024

Language

EN