轻量级视觉变换器实现杂草分割与性别分类实时推理
WeedRepFormer: Reparameterizable Vision Transformers for Real-Time Waterhemp Segmentation and Gender Classification
- 全架构重参数化设计，训练时强、推理时快
- 359万参数下分割mIoU达92.18%，性别分类准确率81.91%
- 适合田间实时智能除草系统部署
我们提出WeedRepFormer，一种轻量级多任务视觉变换器，用于同时实现豚草的分割与性别分类。现有农业模型常难以在细粒度生物特征提取与实时部署效率之间取得平衡。为此，WeedRepFormer在整套架构中系统性地引入结构重参数化——包括视觉变换器主干、轻量级R-ASPP解码器以及新型可重参数化分类头——以解耦训练时的容量与推理时的延迟。我们还构建了一个包含23株豚草共10,264帧标注数据的综合性数据集。在该基准上，WeedRepFormer仅用359万参数和3.80 GFLOPs，实现92.18%的分割mIoU与81.91%的性别分类准确率。在108.95 FPS下，其分类准确率比当前最优iFormer-T高出4.40%，参数量减少1.9倍，同时保持了优异的分割性能。
原文摘要 · Abstract (English)
We present WeedRepFormer, a lightweight multi-task Vision Transformer designed for simultaneous waterhemp segmentation and gender classification. Existing agricultural models often struggle to balance the fine-grained feature extraction required for biological attribute classification with the efficiency needed for real-time deployment. To address this, WeedRepFormer systematically integrates structural reparameterization across the entire architecture - comprising a Vision Transformer backbone, a Lite R-ASPP decoder, and a novel reparameterizable classification head - to decouple training-time capacity from inference-time latency. We also introduce a comprehensive waterhemp dataset containing 10,264 annotated frames from 23 plants. On this benchmark, WeedRepFormer achieves 92.18% mIoU for segmentation and 81.91% accuracy for gender classification using only 3.59M parameters and 3.80 GFLOPs. At 108.95 FPS, our model outperforms the state-of-the-art iFormer-T by 4.40% in classification accuracy while maintaining competitive segmentation performance and significantly reducing parameter count by 1.9x.
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。