Název:	Towards Multi-Species Animal Re-Identification Towards Multi-Species Animal Re-Identification
Autoři:	Fruhner, Maik Tapken, Heiko
Citace zdrojového dokumentu:	WSCG 2024: full papers proceedings: 32. International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, p. 137-146.
Datum vydání:	2024
Nakladatel:	Václav Skala - UNION Agency
Typ dokumentu:	konferenční příspěvek conferenceObject
URI:	http://hdl.handle.net/11025/57386
ISSN:	2464–4625 (online) 2464–4617 (print)
Klíčová slova:	reidentifikace;zvířata;hluboké učení;počítačové vidění
Klíčová slova v dalším jazyce:	re-identification;animals;deep learning;computer vision
Abstrakt v dalším jazyce:	Animal Re-Identification (ReID) is a computer vision task that aims to retrieve a query individual from a gallery of known identities across different camera perspectives. It is closely related to the well-researched topic of Person ReID, but offers a much broader spectrum of features due to the large number of animal species. This raises research questions regarding domain generalization from persons to animals and across multiple animal species. In this paper, we present research on the adaptation of popular deep learning-based person ReID algorithms to the animal domain as well as their ability to generalize across species. We introduce two novel datasets for animal ReID. The first one contains images of 376 different wild common toads. The second dataset consists of various species of zoo animals. Subsequently, we optimize various ReID models on these datasets, as well as on 20 datasets published by others, with the objective of evaluating the performance of the models in a non-person domain. Our findings indicate that the domain generalization capabilities of OSNet AIN extend beyond the person ReID task, despite its comparatively small size. This enables us to investigate real-time animal ReID on live video data.
Práva:	© Václav Skala - UNION Agency
Vyskytuje se v kolekcích:	WSCG 2024: Full Papers Proceedings

Všechny záznamy v DSpace jsou chráněny autorskými právy, všechna práva vyhrazena.