Model-Based Matching by Linear Combinations of Prototypes
We describe a method for modeling object classes (such as faces) using 2D example images and an algorithm for matching a model to a novel image. The object class models are "learned'' from example images that we call prototypes. In addition to the images, the pixelwise correspon...
| Main Authors: | , |
|---|---|
| Language: | en_US |
| Published: |
2004
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/7183 |
| _version_ | 1811069727834898432 |
|---|---|
| author | Jones, Michael J. Poggio, Tomaso |
| author_facet | Jones, Michael J. Poggio, Tomaso |
| author_sort | Jones, Michael J. |
| collection | MIT |
| description | We describe a method for modeling object classes (such as faces) using 2D example images and an algorithm for matching a model to a novel image. The object class models are "learned'' from example images that we call prototypes. In addition to the images, the pixelwise correspondences between a reference prototype and each of the other prototypes must also be provided. Thus a model consists of a linear combination of prototypical shapes and textures. A stochastic gradient descent algorithm is used to match a model to a novel image by minimizing the error between the model and the novel image. Example models are shown as well as example matches to novel images. The robustness of the matching algorithm is also evaluated. The technique can be used for a number of applications including the computation of correspondence between novel images of a certain known class, object recognition, image synthesis and image compression. |
| first_indexed | 2024-09-23T08:14:58Z |
| id | mit-1721.1/7183 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T08:14:58Z |
| publishDate | 2004 |
| record_format | dspace |
| spelling | mit-1721.1/71832019-04-09T17:33:40Z Model-Based Matching by Linear Combinations of Prototypes Jones, Michael J. Poggio, Tomaso AI MIT Artificial Intelligence Computer Vision Image Correspondence Deformable Templates Object Recognition We describe a method for modeling object classes (such as faces) using 2D example images and an algorithm for matching a model to a novel image. The object class models are "learned'' from example images that we call prototypes. In addition to the images, the pixelwise correspondences between a reference prototype and each of the other prototypes must also be provided. Thus a model consists of a linear combination of prototypical shapes and textures. A stochastic gradient descent algorithm is used to match a model to a novel image by minimizing the error between the model and the novel image. Example models are shown as well as example matches to novel images. The robustness of the matching algorithm is also evaluated. The technique can be used for a number of applications including the computation of correspondence between novel images of a certain known class, object recognition, image synthesis and image compression. 2004-10-20T20:49:06Z 2004-10-20T20:49:06Z 1996-12-01 AIM-1583 CBCL-139 http://hdl.handle.net/1721.1/7183 en_US AIM-1583 CBCL-139 33 p. 13000011 bytes 1999501 bytes application/postscript application/pdf application/postscript application/pdf |
| spellingShingle | AI MIT Artificial Intelligence Computer Vision Image Correspondence Deformable Templates Object Recognition Jones, Michael J. Poggio, Tomaso Model-Based Matching by Linear Combinations of Prototypes |
| title | Model-Based Matching by Linear Combinations of Prototypes |
| title_full | Model-Based Matching by Linear Combinations of Prototypes |
| title_fullStr | Model-Based Matching by Linear Combinations of Prototypes |
| title_full_unstemmed | Model-Based Matching by Linear Combinations of Prototypes |
| title_short | Model-Based Matching by Linear Combinations of Prototypes |
| title_sort | model based matching by linear combinations of prototypes |
| topic | AI MIT Artificial Intelligence Computer Vision Image Correspondence Deformable Templates Object Recognition |
| url | http://hdl.handle.net/1721.1/7183 |
| work_keys_str_mv | AT jonesmichaelj modelbasedmatchingbylinearcombinationsofprototypes AT poggiotomaso modelbasedmatchingbylinearcombinationsofprototypes |