Peng LiuDeep convolutional neural networks have been successfully applied to many image-processing problems in recent works. Popular network architectures often add additional operations and connections to the standard architecture to enable training deeper networks. To achieve accurate results in practice, a large number of trainable parameters are often required. Here, we introduce a network architecture based on using dilated convolutions to capture features at different image scales and densely connecting all feature maps with each other. The resulting architecture is able to achieve accurate results with relatively few parameters and consists of a single set of operations, making it easier to implement, train, and apply in practice, and automatically adapts to different problems. We compare results of the proposed network architecture with popular existing architectures for several segmentation problems, showing that the proposed architecture is able to achieve accurate results with fewer parameters, with a reduced risk of overfitting the training data.
Danie ̈lM.Pelt and James.Sethian
https://slidecam-camera.lbl.gov/static/asset/PNAS.pdf
This work uses the dense connection and multiple factors dilated convolution to extract the mixed-scaled features in the same layer.
The idea may have some advantages of extracting mixed scale features. However, it does not provide more comparison study to give us confidence about the effectiveness and novelty of this work.
