video_inference

video_inference copied to clipboard

Author : subarna Tripathi

Enabling co-segmentation of video frames within a single framework. Potentials for test images are kept at CamVid/Result/ and the final results are stored at CamVid/Result/Crf

The following code implements the below papers.

1. Subarna Tripathi, Serge Belongie, Youngbae Hwang, Truong Nguyen Semantic Video Segmentation : Exploring Inference Efficiency ISOCC 2015

2. Subarna Tripathi, Serge Belongie, Truong Nguyen, Beyond Semantic Image Segmentation : Exploring Efficient Inference in Video CVPR 2015 workshop WiCV : Extended Abstract

This software is free ONLY for research purposes. If you want to use any part of the code you should cite this paper.

This code is built on top of the software with below details.

Author: Vibhav Vineet

Contact Information Email : [email protected]

Licence

This software is an environment for pixel-wise labelling problems, designed mainly for object- class segmentation problem and described in detail in

Vibhav Vineet, Jonathan Warrell, Philip H.S. Torr Filter-based Mean-Field Inference for Random Fields with Higher Order Terms and Product Label-Spaces Proceeding of the twelfth European Conference on Computer Vision, 2012.

This software is free ONLY for research purposes. If you want to use any part of the code you should cite this paper.

THIS SOFTWARE IS PROVIDED BY Vibhav Vineet ''AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL Vibhav Vineet BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

This software uses dense pairwise CRF code described in

Philipp Krahenbuhl and Vladlen Koltun Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials Neural Information Processing Systems 2011.

This library is free ONLY for research purposes. If you use this software for research purposes, you should cite this paper in any resulting publication.

This software uses Developer's Image library (DevIL).

Website : http://openil.sourceforge.net/

The image library is free for any purposes.

Potentials for test images are kept at Pascal/Result/ and the final results are stored at Pascal/Result/Crf