Abstract
This paper details the DS@GT team's entry for the AnimalCLEF 2025 re-identification challenge. Our key finding is that the effectiveness of post-hoc metric learning is highly contingent on the initial quality and domain-specificity of the backbone embeddings. We compare a general-purpose model (DINOv2) with a domain-specific model (MegaDescriptor) as a backbone. A K-Nearest Neighbor classifier with robust thresholding then identifies known individuals or flags new ones. While a triplet-learning projection head improved the performance of the specialized MegaDescriptor model by 0.13 points, it yielded minimal gains (0.03) for the general-purpose DINOv2 on averaged BAKS and BAUS. We demonstrate that the general-purpose manifold is more difficult to reshape for fine-grained tasks, as evidenced by stagnant validation loss and qualitative visualizations. This work highlights the critical limitations of refining general-purpose features for specialized, limited-data re-ID tasks and underscores the importance of domain-specific pre-training. The implementation for this work is publicly available at this http URL.
Abstract (translated)
本文详细介绍了DS@GT团队在AnimalCLEF 2025再识别挑战中的参赛方案。我们的主要发现是，后处理度量学习的有效性高度依赖于基础嵌入的初始质量和领域特定性。我们比较了一般用途模型（DINOv2）和专门领域的模型（MegaDescriptor）作为骨干网络的表现。随后使用K-近邻分类器并结合鲁棒阈值来识别已知个体或标记新的个体。虽然三元组学习投影头使专业的MegaDescriptor模型的性能提高了0.13分，但对于一般用途的DINOv2模型，在平均BAKS和BAUS上的改进微乎其微（仅为0.03）。我们展示了通用流形在细粒度任务中更难以重塑，这一点从验证损失停滞以及定性可视化中得以体现。这项工作突显了为专门且数据有限的再识别任务优化通用特征所面临的重要限制，并强调了领域特定预训练的重要性。该工作的实现代码已公开发布于[此处提供URL]。
URL
https://arxiv.org/abs/2509.12353
