DeepLab is a semantic segmentation architecture. First, the input image goes through the network with the use of dilated convolutions. Then the output from the network is bilinearly interpolated and goes through the fully connected CRF to fine tune the result we obtain the final predictions.
Source: Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFsTASK | PAPERS | SHARE |
---|---|---|
Semantic Segmentation | 28 | 36.84% |
Scene Understanding | 3 | 3.95% |
Object Detection | 3 | 3.95% |
Autonomous Driving | 2 | 2.63% |
Medical Image Segmentation | 2 | 2.63% |
Autonomous Vehicles | 2 | 2.63% |
Change detection for remote sensing images | 2 | 2.63% |
Lake Detection | 2 | 2.63% |
Lake Ice Monitoring | 2 | 2.63% |
COMPONENT | TYPE |
|
---|---|---|
![]() |
Structured Prediction | |
![]() |
Convolutions | |
![]() |
Feedforward Networks |