让动物重识别模型在模糊噪声下仍能准辨认，只加0.5%计算量。
DARA: Degradation-Aware Low-Rank Residual Adaptation with Original-to-Corrupted Distillation for Corruption-Robust Animal Re-Identification
- 用低秩残差专家自适应修复退化图像特征，无需标注退化类型。
- 在多个数据集上恢复77%的退化查询准确率损失，接近全量训练效果。
- 适合部署在资源受限设备上的轻量级动物识别系统，泛化性强。
动物重识别依赖细微身份特征，易受模糊、噪声、压缩等视觉退化影响。现有方法通过退化增强训练或像素级修复提升鲁棒性，但未直接修复身份检索空间的偏移。本文将抗退化重识别建模为输入相关的特征空间修复，提出DARA：一种针对紧凑型重识别模型的轻量级改造方案。DARA冻结微调后的主干网络，学习路由的低秩残差专家以适应退化输入嵌入，无需退化类型标注。为稳定修复过程，引入原图到退化图的教师-学生蒸馏，保留个体嵌入和检索关系。在ATRW、FriesianCattle2017、MPDD和SeaStarReID2023上的实验表明，DARA在退化查询检索上优于标准与增强微调方法，可泛化至未见退化与跨域场景，恢复77.0%的退化查询mAP差距至全量退化微调水平，仅增加0.49%参数量和0.05%浮点运算量。
原文摘要 · Abstract (English)
Animal re-identification (Re-ID) relies on fine-grained identity cues that can be disrupted by blur, noise, compression, and other visual degradations. Existing robustness strategies based on degradation-augmented training or pixel-level restoration improve robustness indirectly, but do not explicitly repair shifts in the identity retrieval space. We study corruption-robust animal Re-ID as input-conditioned feature-space repair and introduce DARA, a lightweight retrofit for compact Re-ID models. DARA freezes the fine-tuned backbone and learns routed low-rank residual experts to adapt degraded-input embeddings without corruption-type annotations. To stabilize this adaptive repair, original-to-corrupted distillation uses an original-image teacher to preserve individual embeddings and retrieval relations. Experiments on ATRW, FriesianCattle2017, MPDD, and SeaStarReID2023 show that DARA improves corrupted-query retrieval over standard and augmentation-based fine-tuning, generalizes to unseen corruptions and cross-domain evaluation, and recovers 77.0% of the corrupted-query mAP gap to full corrupted fine-tuning while adding only 0.49% parameters and 0.05% FLOPs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。