Reevaluating Automated Wildlife Species Detection: A Reproducibility Study on a Custom Image Dataset
This work addresses the problem of evaluating reproducibility and generalization in automated wildlife detection for researchers, but it is incremental as it confirms prior findings with minor variations.
This study assessed the reproducibility of a previous method for automated wildlife species detection by reimplementing it on a custom dataset of 900 images across 90 species, achieving 62% accuracy compared to the original 71%, with a macro F1 score of 0.28 indicating variability in performance.
This study revisits the findings of Carl et al., who evaluated the pre-trained Google Inception-ResNet-v2 model for automated detection of European wild mammal species in camera trap images. To assess the reproducibility and generalizability of their approach, we reimplemented the experiment from scratch using openly available resources and a different dataset consisting of 900 images spanning 90 species. After minimal preprocessing, we obtained an overall classification accuracy of 62%, closely aligning with the 71% reported in the original work despite differences in datasets. As in the original study, per-class performance varied substantially, as indicated by a macro F1 score of 0.28,highlighting limitations in generalization when labels do not align directly with ImageNet classes. Our results confirm that pretrained convolutional neural networks can provide a practical baseline for wildlife species identification but also reinforce the need for species-specific adaptation or transfer learning to achieve consistent, high-quality predictions.