Object Recognition/Machine Learning
MENTOR: Victor Eruhimov
We want to improve OpenCV's machine learning.
- Making it easy to run many algorithms on data to assess the performance
- Train models
- You must know C++ well
- You must have experience in Object Recognition
PROJECT: OpenCV Machine Learning
Project Description
Assess and improve the object recognition framework in OpenCV
Demo: object recognition demo using HighGUI
Short term: Samples of object recognition working on open vision datasets
Medium goal: Assess the interface and functionality of OpenCV object recognition toolbox
Long goal: Improve OpenCV object recognition toolbox and develop state of the art recognition models
Milestones/Timeline
What we want
- Assessment and improvement of OpenCV object recognition toolbox on several different problems:
- a) PASCAL 2009
- b) Caltech 101
- c) The ETH Zurich shape database
- d) INRIA person dataset
- During the first phase,
- sample algorithms will be created to assess OpenCV capabilities in these recognition problems.
- Input/output engines will be implemented if necessary, one or several recognition algorithms will be implemented and compared to the state of the art results.
- The convenience of the API will be also assessed.
- Specific projects depend on participants skills and interests, some of possible tasks are:
- Pedestrian detection (sliding window approach with dense descriptors -- HOG, LBP, DAISY -- with SVM and/or cascaded classifier)
- Planar object detection from a single training image (sparse descriptors + RANSAC)
- PASCAL VOC Challenge 2010 (use whatever method available in OpenCV to get the best results on validation set)
- During the second phase,
- the feedback will be used to improve object recognition toolbox both in terms of algorithms and API.
- The project will show what someone can do with a recognition problem having just OpenCV, provide useful feedback for the object recognition toolbox and produce a set of recognition samples.
- Third phase is to develop some useful trained models (for example: humans, face, eye, hand, car ...)