Programme

Workshop Programme

Invited talk by Tim Cootes

Image Segmentation using Statistical Shape and Appearance Models

Abstract:
Many objects of interest in images can be represented as deformed versions of some average structure - for instance faces, bones and many organs in medical images. This talk will describe methods of constructing statistical models of the variation in shape and appearance of such objects from annotated sets of examples. Tim Cootes will give an overview of efficient regression-based techniques for matching such models to new images, including Active Appearance Models and a powerful approach which uses Random Forest regression to vote for the most likely position of each model point. The various methods will be demonstrated on a range of applications, including facial image interpretation and medical image analysis in 2D and 3D images.

 

Proceedings available online (http://www.bmva.org/bmvc/2015/cvppp/index.html)

 

13:30 - 13:40 Opening Remarks

 

13:40 - 14:20 Keynote by Tim Cootes

Image Segmentation using Statistical Shape and Appearance Models, Tim Cootes, University of Manchester

 

14:20 - 15:00 Counting Leaves  

Results from the LCC Challenge: Learning to Count Leaves in Rosette Plants, Mario Valerio Giuffrida, Massimo Minervini, Sotirios Tsaftaris (IMT Lucca)

Counting leaves without ``finger-counting'' by supervised multiscale frequency analysis of depth images from top viewDavid Rousseau (Université Lyon 1), Henricus J. Van de Zedde (Wageningen)

 

15:00-15:30 Coffee break

 

15:30-17:10 Segmentation, Annotation, and Reconstruction 

Results from the LSC Challenge: Utilizing machine learning approaches to improve prediction of leaf counts and individual leaf segmentation of rosette plants, Jean-Michel Pape (IPK-Gatersleben), Christian Klukas (LemnaTec GmbH)

Gaussian process shape models for Bayesian leaf segmentation, Kyle Simek, Kobus Barnard (University of Arizona)

Fuzzy c-means based plant segmentation with distance dependent threshold, Mads Dyrmann (University of Southern Denmark) 

An interactive tool for semi-automated leaf annotation, Massimo Minervini, Mario Valerio Giuffrida, Sotirios Tsaftaris (IMT Lucca) 

3D Surface Reconstruction of Plant Seeds by Volume Carving, Johanna Roussel, Andreas Fischbach, Siegfried Jahnke, Hanno Scharr (Forschungszentrum Jülich)

 

17:10-17:20 Closing Remarks 

Phidas logo

 

EU and FP7 Logos

 

DPPN Logo

 

Some example images from data set

IPPN-Logo