Skip to content

The objective of this thesis is to build a technology which can automatically detect the visually salient region and suppress visually non salient regions of an image.

Notifications You must be signed in to change notification settings

mosharaf13/A-Study-on-Salient-Region-Detection-using-Boundary-and-Color-Cue

Repository files navigation

Visual saliency detection using boundary and color cue

Usage

The branch main points to single image saliency detection algorithm. To run the algorithm do as follows

Prerequisites

Opencv
GCC

On linux debian based systems

cd opencvtest
g++ -g -std=c++11 Source.cpp -o  output $(pkg-config --cflags --libs opencv)
sudo chmod +x output 
./output

The output will generate four images denoting 4 saliency map listed below.

  1. Global saliency map
  2. Color saliency map
  3. Boundary saliency map
  4. Final saliency map

Below diagram denotes relationship between these maps

Results

Initial image

Boundary map Color map Global map

Final saliency map

About

The objective of this thesis is to build a technology which can automatically detect the visually salient region and suppress visually non salient regions of an image.

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages