Multimodal Foundation Models for Zero-shot Animal Species Recognition in Camera Trap Images
This work addresses the need for scalable wildlife tracking with less human labor, though it is incremental as it builds on existing multimodal foundation models.
The authors tackled the problem of costly expert annotations for wildlife monitoring by proposing WildMatch, a zero-shot species classification framework that uses instruction-tuned vision-language models to generate detailed descriptions and match them to an external knowledge base, achieving competitive performance on a new dataset from Colombia.
Due to deteriorating environmental conditions and increasing human activity, conservation efforts directed towards wildlife is crucial. Motion-activated camera traps constitute an efficient tool for tracking and monitoring wildlife populations across the globe. Supervised learning techniques have been successfully deployed to analyze such imagery, however training such techniques requires annotations from experts. Reducing the reliance on costly labelled data therefore has immense potential in developing large-scale wildlife tracking solutions with markedly less human labor. In this work we propose WildMatch, a novel zero-shot species classification framework that leverages multimodal foundation models. In particular, we instruction tune vision-language models to generate detailed visual descriptions of camera trap images using similar terminology to experts. Then, we match the generated caption to an external knowledge base of descriptions in order to determine the species in a zero-shot manner. We investigate techniques to build instruction tuning datasets for detailed animal description generation and propose a novel knowledge augmentation technique to enhance caption quality. We demonstrate the performance of WildMatch on a new camera trap dataset collected in the Magdalena Medio region of Colombia.