用轻量微调让大模型精准分割无人机拍摄的森林地面
Parameter-Efficient Fine-Tuning of Vision Foundation Model for Forest Floor Segmentation from UAV Imagery
- 仅微调少量参数，保持原模型不变以提升效率
- 适配器方法在分割精度上最优，LoRA更适合资源受限设备
- 无需人工标注提示，自动完成树桩、植被等分类
无人飞行器（UAV）在造林与森林监测中应用日益广泛，包括在难以到达区域播撒种子。然而，由于自然变化大、环境参数快速演变以及标注模糊，对林地表面的细致理解仍具挑战。为此，我们采用具备强泛化能力的视觉基础模型Segment Anything Model（SAM），将其用于分割树桩、植被和木屑等林地物体。通过参数高效微调（PEFT）技术，仅更新少量附加参数并固定原始权重。我们调整了SAM的掩码解码器，使其输出对应数据集类别的掩码，实现无需人工提示的自动分割。实验表明，基于适配器的PEFT方法达到最高均交并比（mIoU），而低秩适应（LoRA）因参数更少，成为资源受限无人机平台的轻量替代方案。
原文摘要 · Abstract (English)
Unmanned Aerial Vehicles (UAVs) are increasingly used for reforestation and forest monitoring, including seed dispersal in hard-to-reach terrains. However, a detailed understanding of the forest floor remains a challenge due to high natural variability, quickly changing environmental parameters, and ambiguous annotations due to unclear definitions. To address this issue, we adapt the Segment Anything Model (SAM), a vision foundation model with strong generalization capabilities, to segment forest floor objects such as tree stumps, vegetation, and woody debris. To this end, we employ parameter-efficient fine-tuning (PEFT) to fine-tune a small subset of additional model parameters while keeping the original weights fixed. We adjust SAM's mask decoder to generate masks corresponding to our dataset categories, allowing for automatic segmentation without manual prompting. Our results show that the adapter-based PEFT method achieves the highest mean intersection over union (mIoU), while Low-rank Adaptation (LoRA), with fewer parameters, offers a lightweight alternative for resource-constrained UAV platforms.
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。