Hierarchical fuzzy logic based approach for object tracking
read more
Citations
A multi-view model for visual tracking via correlation filters
Single Object Tracking With Fuzzy Least Squares Support Vector Machine
Shadowed sets of dynamic fuzzy sets
A review and an approach for object detection in images
Visual tracking via exemplar regression model
References
Fuzzy sets
Distinctive Image Features from Scale-Invariant Keypoints
A Computational Approach to Edge Detection
Distinctive Image Features from Scale-Invariant Keypoints
Related Papers (5)
Frequently Asked Questions (16)
Q2. What are the future works in "Hierarchical fuzzy logic based approach for object tracking" ?
Although the presented hierarchical matching approach for multiple object tracking has provided encouraging results, these results lead us to further work with intend to improve robustness, introduce new capabilities and achieve computational efficiency over different image sequences. Further work is intended on the introduction and performance evaluation of different distinctive object properties such as shape, texture and other object descriptors, in order to construct suitable fuzzy sets and introduce new rules in the inference engine.
Q3. What are the main constraints used to deal with the point correspondence problem?
To deal with the point correspondence problem between frames, deterministic constraints such as proximity, maximum velocity and small velocity change could be used.
Q4. What are the possibilities used as visual descriptors of the object appearance?
There’s a myriad of possibilities such as color, shape, texture, size, among others, that can be used as visual descriptors of the object appearance.
Q5. How many objects are removed by the fuzzy algorithm?
After all objects had been processed by the fuzzy algorithm, an update stage is needed to update the Kalman filter and to remove objects with lower confidence degree.
Q6. What is the common method used to measure the similarity between the template and the current target?
The Bhattacharya and Kullback-Leibler distances are commonly employed to measure the similarity between the template and the current target region.
Q7. What is the general operator for the intersection and union of fuzzy sets?
General operators for the intersection and union of fuzzy sets are referred as triangular norms (t-norms) and triangular conorms (t-conorms or s-norms), respectively.
Q8. What is the purpose of the inference engine?
The design of the inference engine should able the process to model the system in order to achieve a good balance between the information provided by the objects kinematic and non kinematic properties.
Q9. What is the alternative approach to detect changes in an image?
An alternative approach to detect changes and, consequently the movement, between two consecutive intensity image frames I(x, y, t) and I(x, y, t − 1) taken at times t and t − 1, respectively, is to perform a pixelwise difference operation.
Q10. What is the proposed methodology for multi feature tracking?
The proposed methodology incorporates hierarchical matching schemes to deal with multi featuretracking and Kalman filters to incorporate the kinematic feature model that increase the processing time.
Q11. How many frames are needed to process?
If there exists more image frames to process, the next frame is analyzed and the cycle is repeated until it reaches the end of the sequence.
Q12. What is the definition of the optical flow of a pixel?
The optical flow of a pixel is a motion vector represented by the translation between a pixel in one frame and its corresponding pixel in the following frame.
Q13. What is the definition of the term KLT?
The KLT is a complete method that provides a solution for two problems in computer vision: the problem of optimal selection of suitable points in an image and the problem of determining the correspondence between points in consecutive frames.
Q14. What are the main characteristics of point based tracking?
Point based tracking approaches are suitable for tracking objects that occupy small regions in an image or they can be represented by several distinctive points.
Q15. What is the reason why the tracking fails?
After frame number 451 the tracking fails since the background denotes higher histogram similarity than the feature histogram due to illumination and object pose changes.
Q16. What is the importance of the update rate for tracking objects?
The update rate is very important to deal with object shape variations, but the higher the update frequency the higher the computational time.