banner about

Cascaded Hierarchical Model (CHM)

Cascaded hierarchical model is an image segmentation framework, which learns contextual information in a hierarchical framework. At each level of the hierarchy, a classifier is trained based on downsampled input image and outputs of previous levels. Our model then incorporates the resulting multi-resolution contextual information into a classifier to segment the input image at original resolution. We repeat this procedure by cascading the hierarchical framework to improve the segmentation accuracy.

download-button


Matlab Codes

  • For scene labeling and edge detection, you should use version 2.0.
  • Precomputed edge detection results on BSDS500 dataset can be found under version 2.0.
  • Version 2.1.330 was released mostly for electron microscopy datasets in collaboration with NCMIR Institute.
  • For information about isntall/usage please check the Readme files included in each package.

Edge Detection

CHM achieves near state-of-the-art performance on Berkeley dataset and outperforms state-of-the-art methods such as SE and SCG on NYU depth dataset v2. The CHM code and corresponding scripts for edge detection for both BSDS500 and NYU dept datasets are included in the download package.

Precomputed testing results and evaluation files for BSDS 500 dataset are also included in the dowload package.

PR BSDS PR NYU
Precision-Recall curves for BSDS500 dataset. Precision-Recall curves for NYU depth v2 dataset.

Scene Labeling

On Stanford background dataset, CHM achieves 82.95% pixel accuracy.

The CHM code and corresponding scripts for scene labeling for the Stanford background dataset are included with the download package.


SBDsamples
Test samples of scene labeling on the Stanford background dataset. First row: input image, Second row: CHM, Third row: CHM with intra-class connection, Fourth row: Groundtruth.

Citation

If you use this package please cite the following:

M. Seyedhosseini, M. Sajjadi, and T. Tasdizen. "Image Segmentation with Cascaded Hierarchical Models and Logistic Disjunctive Normal Networks" In ICCV 2013 (accepted).[pdf, bibtex]

News

  • Version 2.1.330 was released. This new version contains standalone binaries (matlab not required). April 16, 2014.
  • Released the training and testing scripts for Scene labeling and edge detection Feb 4th, 2014.
  • An extended version of the ICCV paper with more results and theoretical insights can be found on arxiv. Feb 4th 2014.
  • Version 2.0 was released (edge filters were added). Dec 29th, 2013.
  • Version 1.0 was released Oct 23rd, 2013.

Contact