A 28M-parameter ConvNeXt-tiny trained with knowledge distillation from BioCLIP2 and 50% target-domain data reaches 89.4% target accuracy, matching the 304M-parameter BioCLIP2 that saw no target data.
An automated light trap to monitor moths (lepidoptera) using computer vision-based tracking and deep learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Bridging Domain Gaps for Fine-Grained Moth Classification Through Expert-Informed Adaptation and Foundation Model Priors
A 28M-parameter ConvNeXt-tiny trained with knowledge distillation from BioCLIP2 and 50% target-domain data reaches 89.4% target accuracy, matching the 304M-parameter BioCLIP2 that saw no target data.