Skip to content

Follows the image occlusion pipeline presented in the paper "Visualising and understanding Convolutional Neural networks" to extract a coarse segmentation map from input image using a pre-trained CNN.

Notifications You must be signed in to change notification settings

akshaychawla/Occlusion-experiments-for-image-segmentation

Repository files navigation

Occlusion experiments for Image Segmentation

Follows the image occlusion pipeline presented in the paper Visualising and understanding Convolutional Neural networks by M. Zeiler et al to extract a coarse segmentation map from input image using a pre-trained CNN.

Look at the included iPython Notebook for the code.

About

Follows the image occlusion pipeline presented in the paper "Visualising and understanding Convolutional Neural networks" to extract a coarse segmentation map from input image using a pre-trained CNN.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published