Using projective invariants for constant time library indexing in model based vision
Projectively invariant shape descriptors allow fast indexing into model libraries, because recognition proceeds without reference to object pose. This paper describes progress in building a large model based vision system which uses many projectively invariant descriptors. We give a brief account of...
| Main Authors: | , , , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
Springer
2012
|
| _version_ | 1811140332900843520 |
|---|---|
| author | Rothwell, CA Zisserman, A Forsyth, DA Mundy, JL |
| author_facet | Rothwell, CA Zisserman, A Forsyth, DA Mundy, JL |
| author_sort | Rothwell, CA |
| collection | OXFORD |
| description | Projectively invariant shape descriptors allow fast indexing into model libraries, because recognition proceeds without reference to object pose. This paper describes progress in building a large model based vision system which uses many projectively invariant descriptors. We give a brief account of these descriptors and then describe the recognition system, giving examples of the invariant techniques working on real images. We demonstrate the ease of model acquisition in our system, where models are generated directly from images. We demonstrate fast recognition without determining object pose or camera parameters. |
| first_indexed | 2024-09-25T04:20:18Z |
| format | Conference item |
| id | oxford-uuid:4ad11859-8a51-40c6-a5cc-30a3fdd085b7 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-09-25T04:20:18Z |
| publishDate | 2012 |
| publisher | Springer |
| record_format | dspace |
| spelling | oxford-uuid:4ad11859-8a51-40c6-a5cc-30a3fdd085b72024-08-08T16:24:06ZUsing projective invariants for constant time library indexing in model based visionConference itemhttp://purl.org/coar/resource_type/c_5794uuid:4ad11859-8a51-40c6-a5cc-30a3fdd085b7EnglishSymplectic ElementsSpringer2012Rothwell, CAZisserman, AForsyth, DAMundy, JLProjectively invariant shape descriptors allow fast indexing into model libraries, because recognition proceeds without reference to object pose. This paper describes progress in building a large model based vision system which uses many projectively invariant descriptors. We give a brief account of these descriptors and then describe the recognition system, giving examples of the invariant techniques working on real images. We demonstrate the ease of model acquisition in our system, where models are generated directly from images. We demonstrate fast recognition without determining object pose or camera parameters. |
| spellingShingle | Rothwell, CA Zisserman, A Forsyth, DA Mundy, JL Using projective invariants for constant time library indexing in model based vision |
| title | Using projective invariants for constant time library indexing in model based vision |
| title_full | Using projective invariants for constant time library indexing in model based vision |
| title_fullStr | Using projective invariants for constant time library indexing in model based vision |
| title_full_unstemmed | Using projective invariants for constant time library indexing in model based vision |
| title_short | Using projective invariants for constant time library indexing in model based vision |
| title_sort | using projective invariants for constant time library indexing in model based vision |
| work_keys_str_mv | AT rothwellca usingprojectiveinvariantsforconstanttimelibraryindexinginmodelbasedvision AT zissermana usingprojectiveinvariantsforconstanttimelibraryindexinginmodelbasedvision AT forsythda usingprojectiveinvariantsforconstanttimelibraryindexinginmodelbasedvision AT mundyjl usingprojectiveinvariantsforconstanttimelibraryindexinginmodelbasedvision |