Wildlife Target Re-Identification Using Self-supervised Learning in Non-Urban Settings
This addresses the problem of data annotation dependency for wildlife researchers, though it is incremental as it applies an existing SSL approach to a new domain.
This study tackled wildlife re-identification by using self-supervised learning with temporal image pairs from camera trap data, eliminating the need for annotated labels. The results showed that self-supervised models are more robust with limited data and outperform supervised features across all downstream tasks.
Wildlife re-identification aims to match individuals of the same species across different observations. Current state-of-the-art (SOTA) models rely on class labels to train supervised models for individual classification. This dependence on annotated data has driven the curation of numerous large-scale wildlife datasets. This study investigates self-supervised learning Self-Supervised Learning (SSL) for wildlife re-identification. We automatically extract two distinct views of an individual using temporal image pairs from camera trap data without supervision. The image pairs train a self-supervised model from a potentially endless stream of video data. We evaluate the learnt representations against supervised features on open-world scenarios and transfer learning in various wildlife downstream tasks. The analysis of the experimental results shows that self-supervised models are more robust even with limited data. Moreover, self-supervised features outperform supervision across all downstream tasks. The code is available here https://github.com/pxpana/SSLWildlife.