This paper implements all the well-known advances in deep learning segmentation methods and make it work together on the CamVid and Gatech datasets.

The proposed network implements the following things:

• Dense blocks
• U-Net architecture

## Dense blocks

The dense block provides a low increase of parameters while keeping a good feature extraction.

## U-Net

The U-Net is well-known to have a good performance on various image segmentation datasets.

## Architecture

The network architecture is a straight U-Net with more convolution stages in each step.

In the following table, $$m$$ represents the number of feature maps output, DB means dense block, TD means transition down and TU means transition up.

### Layers

Here is the internal description of all the blocks defined above.

## Results

Overall the model is as accurate than the other state-of-the-art models while having much fewer parameters (100x less for some of them).

### Issues

Even if the model has fewer parameters than the others models, you should keep in mind that it still has far more layers than a regular U-Net so the memory footprint is huge when you want to train it.

If you run it on theano you should use some THEANO_FLAGS like optimizer_including=fusion to decrease the memory footprint and use a smaller batch_size.

#### Implementations

This original Theano/Lasagne implementations of the paper can be found here FC-DenseNet.

This is my personal Keras/notebook contribution implementation.