MAIS: Memory-Attention for Interactive Segmentation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918017660616704 |
|---|---|
| author | Orbes-Arteaga, Mauricio Lucena, Oeslle Ourselin, Sabastien Cardoso, M. Jorge |
| author_facet | Orbes-Arteaga, Mauricio Lucena, Oeslle Ourselin, Sabastien Cardoso, M. Jorge |
| contents | Interactive medical segmentation reduces annotation effort by refining predictions through user feedback. Vision Transformer (ViT)-based models, such as the Segment Anything Model (SAM), achieve state-of-the-art performance using user clicks and prior masks as prompts. However, existing methods treat interactions as independent events, leading to redundant corrections and limited refinement gains. We address this by introducing MAIS, a Memory-Attention mechanism for Interactive Segmentation that stores past user inputs and segmentation states, enabling temporal context integration. Our approach enhances ViT-based segmentation across diverse imaging modalities, achieving more efficient and accurate refinements. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_07511 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MAIS: Memory-Attention for Interactive Segmentation Orbes-Arteaga, Mauricio Lucena, Oeslle Ourselin, Sabastien Cardoso, M. Jorge Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Interactive medical segmentation reduces annotation effort by refining predictions through user feedback. Vision Transformer (ViT)-based models, such as the Segment Anything Model (SAM), achieve state-of-the-art performance using user clicks and prior masks as prompts. However, existing methods treat interactions as independent events, leading to redundant corrections and limited refinement gains. We address this by introducing MAIS, a Memory-Attention mechanism for Interactive Segmentation that stores past user inputs and segmentation states, enabling temporal context integration. Our approach enhances ViT-based segmentation across diverse imaging modalities, achieving more efficient and accurate refinements. |
| title | MAIS: Memory-Attention for Interactive Segmentation |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2505.07511 |