Understanding the Impact of Training Set Size on Animal Re-identification
This work addresses the challenge of limited training data for wildlife researchers using camera traps, though it is incremental in analyzing existing methods.
The study investigated how training set size affects animal re-identification methods, finding that species-specific traits like intra-individual variance significantly influence data needs, with local feature methods remaining practical for wildlife due to data scarcity.
Recent advancements in the automatic re-identification of animal individuals from images have opened up new possibilities for studying wildlife through camera traps and citizen science projects. Existing methods leverage distinct and permanent visual body markings, such as fur patterns or scars, and typically employ one of two strategies: local features or end-to-end learning. In this study, we delve into the impact of training set size by conducting comprehensive experiments across six different methods and five animal species. While it is well known that end-to-end learning-based methods surpass local feature-based methods given a sufficient amount of good-quality training data, the challenge of gathering such datasets for wildlife animals means that local feature-based methods remain a more practical approach for many species. We demonstrate the benefits of both local feature and end-to-end learning-based approaches and show that species-specific characteristics, particularly intra-individual variance, have a notable effect on training data requirements.