Reevaluating Automated Wildlife Species Detection: A Reproducibility Study on a Custom Image Dataset

TL;DR

The study reproduces Carl et al.'s wildlife-detection approach using a pretrained Inception-ResNet-v2 network on a new 90-species, 900-image dataset to test reproducibility and generalizability with open resources. It reports an overall top-1 accuracy of 0.62, close to the original 0.71 despite dataset differences, and a macro F1 of 0.28, indicating substantial per-species variability due to ImageNet-to-wildlife label mapping. The findings show pretrained CNNs provide a viable baseline for camera-trap annotation but reveal significant generalization gaps without species-specific adaptation or transfer learning. The work also discusses deployment considerations and advocates combining transfer learning with smaller architectures for scalable wildlife monitoring.

Abstract

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.