Geometric Aspects of Visual Object Recognition
This thesis presents there important results in visual object recognition based on shape. (1) A new algorithm (RAST; Recognition by Adaptive Sudivisions of Tranformation space) is presented that has lower average-case complexity than any known recognition algorithm. (2) It is shown, both theor...
| Main Author: | |
|---|---|
| Language: | en_US |
| Published: |
2004
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/7342 |
| _version_ | 1826217154189459456 |
|---|---|
| author | Breuel, Thomas M. |
| author_facet | Breuel, Thomas M. |
| author_sort | Breuel, Thomas M. |
| collection | MIT |
| description | This thesis presents there important results in visual object recognition based on shape. (1) A new algorithm (RAST; Recognition by Adaptive Sudivisions of Tranformation space) is presented that has lower average-case complexity than any known recognition algorithm. (2) It is shown, both theoretically and empirically, that representing 3D objects as collections of 2D views (the "View-Based Approximation") is feasible and affects the reliability of 3D recognition systems no more than other commonly made approximations. (3) The problem of recognition in cluttered scenes is considered from a Bayesian perspective; the commonly-used "bounded-error errorsmeasure" is demonstrated to correspond to an independence assumption. It is shown that by modeling the statistical properties of real-scenes better, objects can be recognized more reliably. |
| first_indexed | 2024-09-23T16:58:53Z |
| id | mit-1721.1/7342 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T16:58:53Z |
| publishDate | 2004 |
| record_format | dspace |
| spelling | mit-1721.1/73422019-04-15T00:40:35Z Geometric Aspects of Visual Object Recognition Breuel, Thomas M. computer vision bouded error point matching 3D objectsrecognition This thesis presents there important results in visual object recognition based on shape. (1) A new algorithm (RAST; Recognition by Adaptive Sudivisions of Tranformation space) is presented that has lower average-case complexity than any known recognition algorithm. (2) It is shown, both theoretically and empirically, that representing 3D objects as collections of 2D views (the "View-Based Approximation") is feasible and affects the reliability of 3D recognition systems no more than other commonly made approximations. (3) The problem of recognition in cluttered scenes is considered from a Bayesian perspective; the commonly-used "bounded-error errorsmeasure" is demonstrated to correspond to an independence assumption. It is shown that by modeling the statistical properties of real-scenes better, objects can be recognized more reliably. 2004-11-19T17:19:47Z 2004-11-19T17:19:47Z 1992-05-01 AITR-1374 http://hdl.handle.net/1721.1/7342 en_US AITR-1374 173 p. 33022903 bytes 26499530 bytes application/postscript application/pdf application/postscript application/pdf |
| spellingShingle | computer vision bouded error point matching 3D objectsrecognition Breuel, Thomas M. Geometric Aspects of Visual Object Recognition |
| title | Geometric Aspects of Visual Object Recognition |
| title_full | Geometric Aspects of Visual Object Recognition |
| title_fullStr | Geometric Aspects of Visual Object Recognition |
| title_full_unstemmed | Geometric Aspects of Visual Object Recognition |
| title_short | Geometric Aspects of Visual Object Recognition |
| title_sort | geometric aspects of visual object recognition |
| topic | computer vision bouded error point matching 3D objectsrecognition |
| url | http://hdl.handle.net/1721.1/7342 |
| work_keys_str_mv | AT breuelthomasm geometricaspectsofvisualobjectrecognition |