WeedRepFormer: Reparameterizable Vision Transformers for Real-Time Waterhemp Segmentation and Gender Classification
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911357601841152 |
|---|---|
| author | Sarker, Toqi Tahamid Islam, Taminul Ahmed, Khaled R. Rankrape, Cristiana Bernardi Creager, Kaitlin E. Gage, Karla |
| author_facet | Sarker, Toqi Tahamid Islam, Taminul Ahmed, Khaled R. Rankrape, Cristiana Bernardi Creager, Kaitlin E. Gage, Karla |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_03431 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | WeedRepFormer: Reparameterizable Vision Transformers for Real-Time Waterhemp Segmentation and Gender Classification Sarker, Toqi Tahamid Islam, Taminul Ahmed, Khaled R. Rankrape, Cristiana Bernardi Creager, Kaitlin E. Gage, Karla Computer Vision and Pattern Recognition 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. |
| title | WeedRepFormer: Reparameterizable Vision Transformers for Real-Time Waterhemp Segmentation and Gender Classification |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.03431 |