Space Efficient 3D Model Indexing
We show that we can optimally represent the set of 2D images produced by the point features of a rigid 3D model as two lines in two high-dimensional spaces. We then decribe a working recognition system in which we represent these spaces discretely in a hash table. We can access this table at r...
| Main Author: | |
|---|---|
| Language: | en_US |
| Published: |
2004
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/5960 |
| _version_ | 1826196114276089856 |
|---|---|
| author | Jacobs, David W. |
| author_facet | Jacobs, David W. |
| author_sort | Jacobs, David W. |
| collection | MIT |
| description | We show that we can optimally represent the set of 2D images produced by the point features of a rigid 3D model as two lines in two high-dimensional spaces. We then decribe a working recognition system in which we represent these spaces discretely in a hash table. We can access this table at run time to find all the groups of model features that could match a group of image features, accounting for the effects of sensing error. We also use this representation of a model's images to demonstrate significant new limitations of two other approaches to recognition: invariants, and non- accidental properties. |
| first_indexed | 2024-09-23T10:21:23Z |
| id | mit-1721.1/5960 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T10:21:23Z |
| publishDate | 2004 |
| record_format | dspace |
| spelling | mit-1721.1/59602019-04-10T17:24:25Z Space Efficient 3D Model Indexing Jacobs, David W. object recognition indexing invariants non-accidentalsproperties hashing space efficiency We show that we can optimally represent the set of 2D images produced by the point features of a rigid 3D model as two lines in two high-dimensional spaces. We then decribe a working recognition system in which we represent these spaces discretely in a hash table. We can access this table at run time to find all the groups of model features that could match a group of image features, accounting for the effects of sensing error. We also use this representation of a model's images to demonstrate significant new limitations of two other approaches to recognition: invariants, and non- accidental properties. 2004-10-04T14:24:11Z 2004-10-04T14:24:11Z 1992-02-01 AIM-1353 http://hdl.handle.net/1721.1/5960 en_US AIM-1353 23 p. 2278295 bytes 1790124 bytes application/postscript application/pdf application/postscript application/pdf |
| spellingShingle | object recognition indexing invariants non-accidentalsproperties hashing space efficiency Jacobs, David W. Space Efficient 3D Model Indexing |
| title | Space Efficient 3D Model Indexing |
| title_full | Space Efficient 3D Model Indexing |
| title_fullStr | Space Efficient 3D Model Indexing |
| title_full_unstemmed | Space Efficient 3D Model Indexing |
| title_short | Space Efficient 3D Model Indexing |
| title_sort | space efficient 3d model indexing |
| topic | object recognition indexing invariants non-accidentalsproperties hashing space efficiency |
| url | http://hdl.handle.net/1721.1/5960 |
| work_keys_str_mv | AT jacobsdavidw spaceefficient3dmodelindexing |