8.19论文阅读
文章目录
- Graph-Segmenter: Graph Transformer with Boundary-aware Attention for Semantic Segmentation
- 方法
- SCSC: Spatial Cross-scale Convolution Module to Strengthen both CNNs and Transformers
- 方法
- Deformable Mixer Transformer with Gating for Multi-Task Learning of Dense Prediction
- 方法
Graph-Segmenter: Graph Transformer with Boundary-aware Attention for Semantic Segmentation
方法
基于图的transformer,
有一定的参考价值
SCSC: Spatial Cross-scale Convolution Module to Strengthen both CNNs and Transformers
方法
就是层级思想
Deformable Mixer Transformer with Gating for Multi-Task Learning of Dense Prediction
方法
代码地址
他的这个是混合注意力