Fully convolutional networks for semantic segmentation

1. How a CNN can be used to do semantic segmentation?
2. What is FCN in image segmentation?
3. Can we use CNN for segmentation?
4. Is ResNet fully convolutional?

How a CNN can be used to do semantic segmentation?

R-CNN (Regions with CNN feature) is one representative work for the region-based methods. It performs the semantic segmentation based on the object detection results. To be specific, R-CNN first utilizes selective search to extract a large quantity of object proposals and then computes CNN features for each of them.

What is FCN in image segmentation?

Fully Convolutional Network

Fully Convolutional Networks, or FCNs, are an architecture used mainly for semantic segmentation. They employ solely locally connected layers, such as convolution, pooling and upsampling. Avoiding the use of dense layers means less parameters (making the networks faster to train).

Can we use CNN for segmentation?

Recently, the two main approaches to image segmentation are based on convolutional neural networks (CNN) and superpixels. Superpixel is an approach that divides an image into regions (called superpixels) with similar properties, such as color, texture, and brightness.

Is ResNet fully convolutional?

FCN-ResNet is constructed by a Fully-Convolutional Network model, using a ResNet-50 or a ResNet-101 backbone.