Bridging Domain Gaps for Fine-Grained Moth Classification Through Expert-Informed Adaptation and Foundation Model Priors
This work addresses the challenge of insect monitoring for ecologists by providing an efficient classification system, though it is incremental as it builds on existing foundation models and distillation techniques.
The paper tackled the problem of accurately classifying moth species from noisy field images by proposing a lightweight method that combines expert-labelled data and knowledge distillation from the BioCLIP2 foundation model, achieving comparable accuracy with reduced computational cost on 101 Danish moth species.
Labelling images of Lepidoptera (moths) from automated camera systems is vital for understanding insect declines. However, accurate species identification is challenging due to domain shifts between curated images and noisy field imagery. We propose a lightweight classification approach, combining limited expert-labelled field data with knowledge distillation from the high-performance BioCLIP2 foundation model into a ConvNeXt-tiny architecture. Experiments on 101 Danish moth species from AMI camera systems demonstrate that BioCLIP2 substantially outperforms other methods and that our distilled lightweight model achieves comparable accuracy with significantly reduced computational cost. These insights offer practical guidelines for the development of efficient insect monitoring systems and bridging domain gaps for fine-grained classification.