Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 43616 (19)
- University: Sharif University Of Technology
- Department: Computer Engineering
- Advisor(s): Jamzad, Mansour
- Abstract:
- Tracking is one of the old and still not thoroughly solved problems in machine vision. Its importance lies on its many applications. These applications vary from security surveillance to examining the motion pattern of atomic particles. There is not a tracker which has acceptable results in all situations, yet. A tracker faces many difficulties such as change in illumination and occlusion. In past, tracking was done by using filters or optical flows. By use of the advances in machine learning and pattern recognition, many models have been proposed to accomplish tracking by using these new learning methods. In this dissertation, we proposed a new tracking method which utilizes sparse representation to overcome the tracking challenges. The sparse representation is selected because it has been proved that natural images have sparse representation. The main idea of sparse representation is to use minimum number of variables to reconstruct a signal.The method which is proposed here, uses a combination of generative and discriminative models, thus, employs both reconstruction power from generative models and discrimination power from discriminative ones. Also, because the tracking problem is online in its nature, the proposed method is online too. The advantage of the online model is that the model updates based on the information from each new frame, thus, the model is more robust to changes in object. Moreover, to prevent the accumulation of errors after updating the model, a generative model has been used. The proposed method is compared with some other methods. These methods use generative models, discriminative models, sparse representation, etc. The experimental results show the superiority of our method in terms of accuracy.
- Keywords:
- Sparse Representation ; Machine Learning ; Visual Tracking ; Dictionary Learning ; Generating Model ; Discriminative Model
Digital Object List
-
محتواي پايان نامه
- view
Bookmark