meta data for this page
  •  

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
mvpr:thesisprojects [2011/05/19 12:51]
jkamarai
mvpr:thesisprojects [2013/01/30 10:59] (current)
jkamarai [Image Alignment Using Pairwise Matching]
Line 11: Line 11:
 ===== Available topics ===== ===== Available topics =====
  
 +==== Optimal Codebook for Visual Bag of Words ====
  
 +Visual Bag-of-Words has become one of the key tools for "​google kind of" image based search. In this project you play with the existing code and data and you writen a machine learning algorithm that will iteratively or randomly generate, test and search optimal codebooks for the BoW based image matching. In particular, the codebooks based on linear filters will be considered.
 +
 +  * C/C++ and/or Matlab skills are required.
 +  * Supervisor: Prof Joni Kamarainen
 +  * Please contact the supervisor for more details.
 +
 +==== Image Alignment Using Pairwise Matching ====
 +
 +The main idea of this project is to extend our previous image alignment method using single global seed, to align images pairwise and then build a tree structure which can
 +align each image to any other image via the tree paths. For more details see our BMVC [[http://​www2.it.lut.fi/​mvpr/​data/​Lankinen-bmvc2011.pdf|paper]].
 +
 +  * C/C++ and/or Matlab skills are required.
 +  * Supervisor: Prof Joni Kamarainen
 +  * Please contact the supervisor for more details.
 +
 +==== Model Predictive Visual Servoing ====
 +
 +Visual servoing means the real-time control of robots using vision (images). Model predictive control is a modern control paradigm which uses optimization to choose optimal control decisions by predicting a future evolution of a system and optimizing a control objective for the prediction horizon. Using model predictive control with visual servoing is still in its infancy. In this project you can learn about state-of-the-art of visual control and develop the art even further. ​
 +
 +  * The thesis should review existing literature on visual servoing using model predictive control. An existing robotic system should also be developed further to perform model predictive visual servoing.
 +  * The implementation can be done on MVPR's MELFA robot arm.
 +  * C++ skills required. Matlab skills might be useful in prototyping.
 +  * Supervisor: Prof Ville Kyrki
 +
 +==== Learning Grasp Affordances from Vision ====
 +To grasp an object, a robot needs to know where to place its fingers. The set of good finger placements for a particular object are called grasp affordances. Determining grasp affordances from visual input for unknown objects has gained much interest in robotics research community recently, as this would allow robots to operate in normal household environments. ​
 +  * The thesis should review the recent work on learning grasp affordances from vision and implement an example system using a state-of-the-art approach.
 +  * The implementation can be done on MVPR's MELFA robot arm, Weiss robotics/​Schunk gripper, and Kinect sensor.
 +  * C++ and/or Matlab skills are required.
 +  * Supervisor: Prof Ville Kyrki
 +  * Please contact the supervisor for more details.
 +
 +==== 3D Interest Points from Stereo Images ====
 +
 +Interest points have been the hot topic of computer vision for some time and they are the low level operators for image based search engines.
 +In this project you will step on a cutting-edge topic by developing such interest points further - to be used with a stereo pair images. This means
 +that on the low level you utilise existing interest point detectors, but then you should select only those interest points which match between the
 +left and right view and then add 3D information to them (depth). In this project you learn about stereo imaging and state-of-the-art methods for image-based search.
 +
 +  * C/C++ and/or Matlab skills are required.
 +  * Supervisor: Prof Joni Kamarainen
 +  * Please contact the supervisor for more details.
 +
 +==== What Really is Important in Objects ====
 +
 +This work is based on the hypothesis that there are certain local features which are more important than other. Those features can be found from many natural objects and they appear similarly in objects of a same class (car, face, etc.) In this work you will study such important features and make a detector which automatically finds them. First you will use our existing framework to select points which appear similarly in other examples of the same
 +class and then you devise a detector especially for these points. ​
 +
 +  * C++ and/or Matlab skills are required.
 +  * Supervisor: Prof Joni Kamarainen
 +  * Please contact the supervisor for more details.
 +
 +==== Symbolic Description of 3D Objects ====
 +
 +In this work, you learn some cutting edge technologies. You will learn about local features which are used in modern image based search technologies. Moreover, you will learn about stereo images which can now be produced by off-the-shelf cameras. Your task is to describe a 3D object, captured by a stereo camera, using local symbols - a kind of "3D interest points"​.
 +
 +  * C++ and/or Matlab skills are required.
 +  * Supervisor: Prof Joni Kamarainen
 +  * Please contact the supervisor for more details.
 +
 +==== Detection of Things from Images ====
 +
 +How to detect that there is a "​thing"​ or things in an image. A thing is something which will be interesting for humans and/or important for automatic image retrieval. Humans, cars, buildings are all "​things",​ but how to automatically detect them. Can you make an object specific detector or even a detector for all things. That will be experimentally investigated in your work. In this work you can do play with a cutting edge problem that also interests certain big companies at the moment.
 +
 +  * C++ and/or Matlab skills are required.
 +  * Supervisor: Prof Joni Kamarainen
 +  * Please contact the supervisor for more details.
 +
 +==== Probabilistic Robotic Manipulation ====
 +
 +One of the difficulties in using robots in home-like environments is the complexity and difficulty of observing such environments. Thus, adaptation to the environment and toleration of uncertainty in the robot'​s knowledge would be very valuable. In effect, the uncertainty is usually modeled using probabilistic models. Robotic manipulation in uncertain conditions has recently gained lots of attention from researchers world-wide and we in Lappeenranta are at the forefront of the research. ​
 +  * The thesis should review the recent work on manipulation under uncertainty and implement an example system using a state-of-the-art approach.
 +  * The implementation can be done on MVPR's MELFA robot arm, Weiss robotics/​Schunk gripper, and Point Grey stereo camera.
 +  * C++ and/or Matlab skills are required.
 +  * Working knowledge of basic probability theory is required (e.g., completing Pattern Recognition course).
 +  * Supervisor: Prof Ville Kyrki
 +  * Please contact the supervisor for more details.
 +
 +==== Lost in Probabilities ====
 +Probability theory and statistics provide the fundamental background for solving the most difficult problems in computer vision, pattern recognition,​ machine learning and engineering in general. Do you still feel uncomfortable while encountering probabilities?​ Do you, however, want to master probabilities and important concepts related them? In this project you will learn about the basic things with probabilities. You will learn what is likelihood, study how to combine likelihoods,​ how to transform likelihoods to probabilities and play with the concept of "​probability score" developed in our laboratory. How the likelihoods can be reliably converted to probability scores and are the probability scores probabilities themselves? Your main task is to briefly review the basic probability theory, spot the most important concepts, review related literature for the selected concepts, explain the main results and program simple examples to explain and verify the theory. The main emphasis is on single and multiple Gaussian densities which have been found very efficient methods for computer vision problems in our laboratory. If you enjoy mathematics and enjoy learning new and powerful things, this could be your project.
 +  * Requires Matlab programming (examples)
 +  * Supervisor: Prof Joni Kämäräinen
 +  * Please contact the supervisor for more details
  
 ==== Robust Object Class Descriptors ==== ==== Robust Object Class Descriptors ====
Line 21: Line 105:
   * Requires programming   * Requires programming
   * Supervisor: Prof Joni Kämäräinen   * Supervisor: Prof Joni Kämäräinen
-  * Please contact the supervisor for more details 
- 
-==== Visual Object Categorisation with Very Simple Features ==== 
-Visual object categorisation means systems which take can automatically detect which learned objects 
-appear in input images. This is a hot topic in computer vision and artificial intelligence - 
-see [[http://​www.vision.ee.ethz.ch/​~bleibe/​teaching/​tutorial-aaai08/​]] for a brief tutorial. 
- 
-People are using fancier and fancier methods to solve this problem, but in this project we go 
-back to very basics and utilise the most simplest approaches to solve the problem. My claim is 
-that these simple methods, when properly utilised, are not actually that bad at all. As a result 
-you will learn how the new applications,​ such as Google image search, really work. 
-  * Experiments implemented on Matlab (good knowledge on any programming language is sufficient) 
-  * Supervisor: Prof Joni Kamarainen 
   * Please contact the supervisor for more details   * Please contact the supervisor for more details
  
Line 57: Line 128:
   * Supervisor: Prof Ville Kyrki   * Supervisor: Prof Ville Kyrki
   * Please contact the supervisor for more details.   * Please contact the supervisor for more details.
 +
 +==== Camera-Projector System Calibration Using a Flat Display ====
 +In this work you will learn how to calibrate a digital camera and a projector. You will implement a GUI based calibration system, which
 +uses a display instead of a physical calibration pattern. You will learn about geometry between the world, camera and projector.
 +  * Requires programming (mainly Matlab)
 +  * Supervisor: Prof Joni Kämäräinen
 +  * Please contact the supervisor for more details
  
 ===== Reserved topics ===== ===== Reserved topics =====
 +
 +==== Visual Object Categorisation with Very Simple Features (* reserved Sept 2011 *) ====
 +Visual object categorisation means systems which take can automatically detect which learned objects
 +appear in input images. This is a hot topic in computer vision and artificial intelligence -
 +see [[http://​www.vision.ee.ethz.ch/​~bleibe/​teaching/​tutorial-aaai08/​]] for a brief tutorial.
 +
 +People are using fancier and fancier methods to solve this problem, but in this project we go
 +back to very basics and utilise the most simplest approaches to solve the problem. My claim is
 +that these simple methods, when properly utilised, are not actually that bad at all. As a result
 +you will learn how the new applications,​ such as Google image search, really work.
 +  * Experiments implemented on Matlab (good knowledge on any programming language is sufficient)
 +  * Supervisor: Prof Joni Kamarainen
 +  * Please contact the supervisor for more details
  
 ==== Assisted 3D reconstruction from a single or multiple images (*reserved May 2011*) ==== ==== Assisted 3D reconstruction from a single or multiple images (*reserved May 2011*) ====
Line 65: Line 156:
   * Supervisor: Prof Joni Kämäräinen   * Supervisor: Prof Joni Kämäräinen
   * Please contact the supervisor for more details   * Please contact the supervisor for more details
- 
-==== 2D/3D Tracking Using Wii Remote (*reserved Oct 2010*) ==== 
- 
-In this task, you must use Wii remote functionality to provide coordinates of up-to 4 "​blobs"​ coming from IR sources. To accomplish the task you need find proper IR sources (LEDs?), read bluetooth data from Wii Remote, and display Wii track data on a computer display (using, e.g., OpenGL). Computing and hardware resources will be provided by the lab. 
- 
-  * Implementation will be done using Linux (Ubuntu) environment and C/C++ language. Great demos on similar works for Windows can be found from: [[http://​johnnylee.net/​projects/​wii/​]] 
-  * Supervisor: Joni Kamarainen 
-  * Contact supervisor for more details ​ 
  
 ==== System Identification of Bacteriorhodopsin-Based Photoelectric Sensors (** reserved Nov 2010 **) ==== ==== System Identification of Bacteriorhodopsin-Based Photoelectric Sensors (** reserved Nov 2010 **) ====
Line 80: Line 163:
   * Supervisor: Assoc. Prof. Lasse Lensu   * Supervisor: Assoc. Prof. Lasse Lensu
   * Please contact the supervisor for more details   * Please contact the supervisor for more details
- 
-==== Image Color Transformations (** reserved Oct 2010 **) ==== 
-Have you ever thought why colours in two different images from the same scene are so different. In this project you will find the answer and you will implement several known methods to "​register"​ colours of one image to another. Moreover, you can play with the colour transformations and generate weird looking absurd images from any input images. Moreover, colour normalisation is very important, for example, in medical imaging applications. 
- 
-  * Application:​ Image Color Calibration,​ colour normalisation of medical images. 
-  * Supervisor: Prof Joni Kämäräinen 
-  * Please contact the supervisor for more details 
- 
- 
- 
- 
  
 ==== Image Repainting With General Codes (* Reserved May 2010 *) ==== ==== Image Repainting With General Codes (* Reserved May 2010 *) ====