Semantic Segmentation Modules
Deeper Atrous Spatial Pyramid Pooling
Introduced by Emara et al. in LiteSeg: A Novel Lightweight ConvNet for Semantic SegmentationDASPP is a deeper version of the ASPP module (the latter from DeepLabv3) that adds standard 3 × 3 convolution after 3 × 3 dilated convolutions to refine the features and also fusing the input and the output of the DASPP module via short residual connection. Also, the number of convolution filters of ASPP is reduced from 255 to 96 to gain computational performance.
Source: LiteSeg: A Novel Lightweight ConvNet for Semantic Segmentation
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Computational Efficiency | 1 | 25.00% |
| Image Segmentation | 1 | 25.00% |
| Real-Time Semantic Segmentation | 1 | 25.00% |
| Semantic Segmentation | 1 | 25.00% |
1x1 Convolution
Average Pooling
Convolution
Dilated Convolution
Residual Connection
Spatial Pyramid Pooling
