Difference between revisions of "SoC2007 project Feature Descriptor"

From PanoTools.org Wiki
Jump to: navigation, search
(initial version for further expansion and discussion)
 
m (Robust local feature detector and descriptor)
Line 4: Line 4:
 
A detector and descriptor that takes into account the approximately known distortions will have a much higher matching rate, especially when fisheye or wide angle images are used.  
 
A detector and descriptor that takes into account the approximately known distortions will have a much higher matching rate, especially when fisheye or wide angle images are used.  
  
This project should be split into two projects, which could be done by separate students:
 
 
* Implementation of the feature detector and descriptor, and a suitable test suite to verify the correctness of the implementation.
 
* Implementation of the feature detector and descriptor, and a suitable test suite to verify the correctness of the implementation.
  
Line 10: Line 9:
 
A desired result of the projects would be:
 
A desired result of the projects would be:
 
* C or C++ library that implements the detection and description steps.
 
* C or C++ library that implements the detection and description steps.
* C or C++ library for the matching step.
+
* An executable for extracting features from image file.
* Integration of both libraries into [[hugin]] and a standalone executable similar to [[Autopano-sift]] or [[Autopano]]
+
* Test suite to evaluate descriptor on a large amount of images.
 +
* Integration of the library into [[hugin]]
  
 
== Timeline ==
 
== Timeline ==
Line 27: Line 27:
 
* [http://homes.esat.kuleuven.be/%7Etuytelaa/ECCV06tutorial.html Tutorial on local invariant features]
 
* [http://homes.esat.kuleuven.be/%7Etuytelaa/ECCV06tutorial.html Tutorial on local invariant features]
 
* [http://www.mis.informatik.tu-darmstadt.de/Education/Courses/cv/index.html Lecture notes] on feature detection/matching]
 
* [http://www.mis.informatik.tu-darmstadt.de/Education/Courses/cv/index.html Lecture notes] on feature detection/matching]
 +
 +
=== Software ===
 
* Framework for [http://www.robots.ox.ac.uk/~vgg/research/affine/ testing descriptors]
 
* Framework for [http://www.robots.ox.ac.uk/~vgg/research/affine/ testing descriptors]
 
=== Software and libraries ===
 
 
  
 
Mentor: Pablo d'Angelo, Herbert Bay, ?
 
Mentor: Pablo d'Angelo, Herbert Bay, ?
  
 
License: GPL
 
License: GPL

Revision as of 08:46, 16 March 2007

Robust local feature detector and descriptor

Goal: Robust matching of features between multiple images using a Hessian-based detector and a suitable descriptor. A detector and descriptor that takes into account the approximately known distortions will have a much higher matching rate, especially when fisheye or wide angle images are used.

  • Implementation of the feature detector and descriptor, and a suitable test suite to verify the correctness of the implementation.

Deliverables

A desired result of the projects would be:

  • C or C++ library that implements the detection and description steps.
  • An executable for extracting features from image file.
  • Test suite to evaluate descriptor on a large amount of images.
  • Integration of the library into hugin

Timeline

Required/available ressources

Required knowledge or interest in:

  • signal or image processing background
  • C or C++ development skills.
  • Matlab or octave

Ressources

Literature

Literature about feature detection

Software

Mentor: Pablo d'Angelo, Herbert Bay, ?

License: GPL