Error weighted classifier combination for multi-modal human identification
In this paper we describe a technique of classifier combination used in a human identification system. The system integrates all available features from multi-modal sources within a Bayesian framework. The framework allows representinga class of popular classifier combination rules and methods withi...
| Main Authors: | , , |
|---|---|
| Language: | en_US |
| Published: |
2005
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/30590 |
| _version_ | 1826189407017762816 |
|---|---|
| author | Ivanov, Yuri Serre, Thomas Bouvrie, Jacob |
| author_facet | Ivanov, Yuri Serre, Thomas Bouvrie, Jacob |
| author_sort | Ivanov, Yuri |
| collection | MIT |
| description | In this paper we describe a technique of classifier combination used in a human identification system. The system integrates all available features from multi-modal sources within a Bayesian framework. The framework allows representinga class of popular classifier combination rules and methods within a single formalism. It relies on a Âper-class measure of confidence derived from performance of each classifier on training data that is shown to improve performance on a synthetic data set. The method is especially relevant in autonomous surveillance setting where varying time scales and missing features are a common occurrence. We show an application of this technique to the real-world surveillance database of video and audio recordings of people collected over several weeks in the office setting. |
| first_indexed | 2024-09-23T08:14:33Z |
| id | mit-1721.1/30590 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T08:14:33Z |
| publishDate | 2005 |
| record_format | dspace |
| spelling | mit-1721.1/305902019-04-09T17:31:29Z Error weighted classifier combination for multi-modal human identification Ivanov, Yuri Serre, Thomas Bouvrie, Jacob AI classifier combination face recognition identification multi-modal In this paper we describe a technique of classifier combination used in a human identification system. The system integrates all available features from multi-modal sources within a Bayesian framework. The framework allows representinga class of popular classifier combination rules and methods within a single formalism. It relies on a Âper-class measure of confidence derived from performance of each classifier on training data that is shown to improve performance on a synthetic data set. The method is especially relevant in autonomous surveillance setting where varying time scales and missing features are a common occurrence. We show an application of this technique to the real-world surveillance database of video and audio recordings of people collected over several weeks in the office setting. 2005-12-22T02:44:17Z 2005-12-22T02:44:17Z 2005-12-14 MIT-CSAIL-TR-2005-081 AIM-2005-035 CBCL-258 http://hdl.handle.net/1721.1/30590 en_US Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory 7 p. 22108540 bytes 952178 bytes application/postscript application/pdf application/postscript application/pdf |
| spellingShingle | AI classifier combination face recognition identification multi-modal Ivanov, Yuri Serre, Thomas Bouvrie, Jacob Error weighted classifier combination for multi-modal human identification |
| title | Error weighted classifier combination for multi-modal human identification |
| title_full | Error weighted classifier combination for multi-modal human identification |
| title_fullStr | Error weighted classifier combination for multi-modal human identification |
| title_full_unstemmed | Error weighted classifier combination for multi-modal human identification |
| title_short | Error weighted classifier combination for multi-modal human identification |
| title_sort | error weighted classifier combination for multi modal human identification |
| topic | AI classifier combination face recognition identification multi-modal |
| url | http://hdl.handle.net/1721.1/30590 |
| work_keys_str_mv | AT ivanovyuri errorweightedclassifiercombinationformultimodalhumanidentification AT serrethomas errorweightedclassifiercombinationformultimodalhumanidentification AT bouvriejacob errorweightedclassifiercombinationformultimodalhumanidentification |