EcologyInternational Journal of Computer VisionPeer-reviewed
WildCLIP: Scene and animal attribute retrieval from camera trap data with domain-adapted vision-language models
Valentin Gabeff, Marc Rußwurm, Devis Tuia, Alexander Mathis
WildCLIP retrieves camera trap images based on environmental and ecological attributes using a domain-adapted vision-language model.
40 citationstop 4%