用混合分歧与多样性策略，仅用2.3%标注数据实现高精度生物声学事件检测。
Hybrid Disagreement-Diversity Active Learning for Bioacoustic Sound Event Detection
- 结合委员会投票分歧与样本多样性，动态筛选最有价值的待标注数据。
- 冷启动下达到68% mAP，仅需2.3%标注数据，接近全监督75%的性能。
- 对稀有物种和冷启动场景表现优异，适合濒危物种监测应用。
生物声学事件检测（BioSED）对生物多样性保护至关重要，但模型开发面临标注数据有限、事件稀疏、物种多样性和类别不平衡等挑战。为在有限标注预算下高效应对，本文采用融合委员会投票分歧与多样性分析的‘先分歧后远距遍历’（MFFT）主动学习方法，并专门优化了一个现有BioSED数据集以评估主动学习算法。实验表明，MFFT在冷启动条件下实现68%的mAP，暖启动时达71%（接近全监督75%的mAP），仅使用2.3%的标注数据。尤其在冷启动和稀有物种检测中表现突出，展现出实际应用价值。
原文摘要 · Abstract (English)
Bioacoustic sound event detection (BioSED) is crucial for biodiversity conservation but faces practical challenges during model development and training: limited amounts of annotated data, sparse events, species diversity, and class imbalance. To address these challenges efficiently with a limited labeling budget, we apply the mismatch-first farthest-traversal (MFFT), an active learning method integrating committee voting disagreement and diversity analysis. We also refine an existing BioSED dataset specifically for evaluating active learning algorithms. Experimental results demonstrate that MFFT achieves a mAP of 68% when cold-starting and 71% when warm-starting (which is close to the fully-supervised mAP of 75%) while using only 2.3% of the annotations. Notably, MFFT excels in cold-start scenarios and with rare species, which are critical for monitoring endangered species, demonstrating its practical value.
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。