Skip to content

Perform image segmentation and background removal in javascript using canvas element and computer vision superpixel algorithms, and convnetjs (experimental)

Notifications You must be signed in to change notification settings

wuhang/ImageSegmentation

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Image Eraser by Akshay Bhat

http://eraseimage.com/ Image Eraser allows users to perform image segmentation inside browser using a vector editor (FabricJS) and JS implementations of superpixel algorithms.

Very early experimental results using ConvnetJS (ordinary NN not deep net) and support for Zoom In/Out using scrollwheel/touchpad at http://www.eraseimage.com/Experiment

Using SLIC superpixels

Segmentation

Pixelwise prediction using ConvnetJS (with Zoom support)

ConvnetJS

Following libraries & templates are used:

  1. https://almsaeedstudio.com/
  2. https://github.com/kyamagu/js-segment-annotator/blob/master/LICENSE
  3. http://fabricjs.com/kitchensink/
  4. http://inspirit.github.io/jsfeat/
  5. http://linkedin.github.io/hopscotch/
  6. https://github.com/karpathy/convnetjs
  7. http://ace.c9.io/
  8. https://github.com/axemclion/seamcarving

License:
Copyright 2015, Cornell University. All Rights Reserved.

About

Perform image segmentation and background removal in javascript using canvas element and computer vision superpixel algorithms, and convnetjs (experimental)

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages

  • JavaScript 88.9%
  • CSS 5.8%
  • HTML 5.2%
  • Python 0.1%