On the Verification of Hypothesized Matches in Model-Based Recognition
In model-based recognition, ad hoc techniques are used to decide if a match of data to model is correct. Generally an empirically determined threshold is placed on the fraction of model features that must be matched. We rigorously derive conditions under which to accept a match, relating the...
| Main Authors: | , |
|---|---|
| Language: | en_US |
| Published: |
2004
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/6028 |
| _version_ | 1826202981381439488 |
|---|---|
| author | Grimson, W. Eric L. Huttenlocher, Daniel P. |
| author_facet | Grimson, W. Eric L. Huttenlocher, Daniel P. |
| author_sort | Grimson, W. Eric L. |
| collection | MIT |
| description | In model-based recognition, ad hoc techniques are used to decide if a match of data to model is correct. Generally an empirically determined threshold is placed on the fraction of model features that must be matched. We rigorously derive conditions under which to accept a match, relating the probability of a random match to the fraction of model features accounted for, as a function of the number of model features, number of image features and the sensor noise. We analyze some existing recognition systems and show that our method yields results comparable with experimental data. |
| first_indexed | 2024-09-23T12:29:13Z |
| id | mit-1721.1/6028 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T12:29:13Z |
| publishDate | 2004 |
| record_format | dspace |
| spelling | mit-1721.1/60282019-04-12T08:28:44Z On the Verification of Hypothesized Matches in Model-Based Recognition Grimson, W. Eric L. Huttenlocher, Daniel P. object recognition search model-based vision In model-based recognition, ad hoc techniques are used to decide if a match of data to model is correct. Generally an empirically determined threshold is placed on the fraction of model features that must be matched. We rigorously derive conditions under which to accept a match, relating the probability of a random match to the fraction of model features accounted for, as a function of the number of model features, number of image features and the sensor noise. We analyze some existing recognition systems and show that our method yields results comparable with experimental data. 2004-10-04T14:36:17Z 2004-10-04T14:36:17Z 1989-05-01 AIM-1110 http://hdl.handle.net/1721.1/6028 en_US AIM-1110 23 p. 3009307 bytes 1200576 bytes application/postscript application/pdf application/postscript application/pdf |
| spellingShingle | object recognition search model-based vision Grimson, W. Eric L. Huttenlocher, Daniel P. On the Verification of Hypothesized Matches in Model-Based Recognition |
| title | On the Verification of Hypothesized Matches in Model-Based Recognition |
| title_full | On the Verification of Hypothesized Matches in Model-Based Recognition |
| title_fullStr | On the Verification of Hypothesized Matches in Model-Based Recognition |
| title_full_unstemmed | On the Verification of Hypothesized Matches in Model-Based Recognition |
| title_short | On the Verification of Hypothesized Matches in Model-Based Recognition |
| title_sort | on the verification of hypothesized matches in model based recognition |
| topic | object recognition search model-based vision |
| url | http://hdl.handle.net/1721.1/6028 |
| work_keys_str_mv | AT grimsonwericl ontheverificationofhypothesizedmatchesinmodelbasedrecognition AT huttenlocherdanielp ontheverificationofhypothesizedmatchesinmodelbasedrecognition |