Improved appearance updating method in multiple instance learning tracking
Multiple instance learning (MIL) tracker becomes recently very popular because of their great success in complex scenes. Dynamically reflecting the appearance changes of the tracked object, the appearance updating plays an important role on tracking. In the original MIL tracker, the appearance model...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2014-04-01
|
| Series: | IET Computer Vision |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/iet-cvi.2013.0006 |
| _version_ | 1827817987301703680 |
|---|---|
| author | Jifeng Ning Wuzhen Shi Shuqin Yang Paul Yanne |
| author_facet | Jifeng Ning Wuzhen Shi Shuqin Yang Paul Yanne |
| author_sort | Jifeng Ning |
| collection | DOAJ |
| description | Multiple instance learning (MIL) tracker becomes recently very popular because of their great success in complex scenes. Dynamically reflecting the appearance changes of the tracked object, the appearance updating plays an important role on tracking. In the original MIL tracker, the appearance model is assumed to obey normal distribution and its updating rule consists of a simple linearly weighted sum of the original and the current target distributions in the current frame. However, this updating method is not proved theoretically. In this work, the authors deduce a novel appearance updating method by estimating the mean and the variance of the sum of two normal distributions being merged in maximum likelihood estimation. The method can be naturally extended to multivariable distributions, useful to track colour object. Experimental results on some benchmark video sequences show that the method achieve higher precision and reliability than the three state‐of‐art trackers. |
| first_indexed | 2024-03-12T00:41:24Z |
| format | Article |
| id | doaj.art-951b14cb9c0a43b082feca7a8a821a3a |
| institution | Directory Open Access Journal |
| issn | 1751-9632 1751-9640 |
| language | English |
| last_indexed | 2024-03-12T00:41:24Z |
| publishDate | 2014-04-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Computer Vision |
| spelling | doaj.art-951b14cb9c0a43b082feca7a8a821a3a2023-09-15T07:13:10ZengWileyIET Computer Vision1751-96321751-96402014-04-018211813010.1049/iet-cvi.2013.0006Improved appearance updating method in multiple instance learning trackingJifeng Ning0Wuzhen Shi1Shuqin Yang2Paul Yanne3College of Information EngineeringNorthwest A&F UniversityYangling712100People's Republic of ChinaCollege of Information EngineeringNorthwest A&F UniversityYangling712100People's Republic of ChinaCollege of Mechanical and Electronic EngineeringNorthwest A&F UniversityYangling712100People's Republic of ChinaCollege of Information EngineeringNorthwest A&F UniversityYangling712100People's Republic of ChinaMultiple instance learning (MIL) tracker becomes recently very popular because of their great success in complex scenes. Dynamically reflecting the appearance changes of the tracked object, the appearance updating plays an important role on tracking. In the original MIL tracker, the appearance model is assumed to obey normal distribution and its updating rule consists of a simple linearly weighted sum of the original and the current target distributions in the current frame. However, this updating method is not proved theoretically. In this work, the authors deduce a novel appearance updating method by estimating the mean and the variance of the sum of two normal distributions being merged in maximum likelihood estimation. The method can be naturally extended to multivariable distributions, useful to track colour object. Experimental results on some benchmark video sequences show that the method achieve higher precision and reliability than the three state‐of‐art trackers.https://doi.org/10.1049/iet-cvi.2013.0006improved appearance updating methodmultiple instance learning trackingMIL trackerappearance modelnormal distributionupdating rule |
| spellingShingle | Jifeng Ning Wuzhen Shi Shuqin Yang Paul Yanne Improved appearance updating method in multiple instance learning tracking IET Computer Vision improved appearance updating method multiple instance learning tracking MIL tracker appearance model normal distribution updating rule |
| title | Improved appearance updating method in multiple instance learning tracking |
| title_full | Improved appearance updating method in multiple instance learning tracking |
| title_fullStr | Improved appearance updating method in multiple instance learning tracking |
| title_full_unstemmed | Improved appearance updating method in multiple instance learning tracking |
| title_short | Improved appearance updating method in multiple instance learning tracking |
| title_sort | improved appearance updating method in multiple instance learning tracking |
| topic | improved appearance updating method multiple instance learning tracking MIL tracker appearance model normal distribution updating rule |
| url | https://doi.org/10.1049/iet-cvi.2013.0006 |
| work_keys_str_mv | AT jifengning improvedappearanceupdatingmethodinmultipleinstancelearningtracking AT wuzhenshi improvedappearanceupdatingmethodinmultipleinstancelearningtracking AT shuqinyang improvedappearanceupdatingmethodinmultipleinstancelearningtracking AT paulyanne improvedappearanceupdatingmethodinmultipleinstancelearningtracking |