Automatic Class-Specific 3D Reconstruction from a Single Image
Our goal is to automatically reconstruct 3D objects from a single image, by using prior 3D shape models of classes. The shape models, defined as a collection of oriented primitive shapes centered at fixed 3D positions, can be learned from a few labeled images for each class. The 3D class model can t...
| मुख्य लेखकों: | , , |
|---|---|
| अन्य लेखक: | |
| प्रकाशित: |
2009
|
| ऑनलाइन पहुंच: | http://hdl.handle.net/1721.1/44615 |
| _version_ | 1826191430553436160 |
|---|---|
| author | Lozano-Perez, Tomas Kaelbling, Leslie Pack Chiu, Han-Pang |
| author2 | Tomas Lozano-Perez |
| author_facet | Tomas Lozano-Perez Lozano-Perez, Tomas Kaelbling, Leslie Pack Chiu, Han-Pang |
| author_sort | Lozano-Perez, Tomas |
| collection | MIT |
| description | Our goal is to automatically reconstruct 3D objects from a single image, by using prior 3D shape models of classes. The shape models, defined as a collection of oriented primitive shapes centered at fixed 3D positions, can be learned from a few labeled images for each class. The 3D class model can then be used to estimate the 3D shape of an object instance, including occluded parts, from a single image. We provide a quantitative evaluation of the shape estimation process on real objects and demonstrate its usefulness in three applications: robot manipulation, object detection, and generating 3D 'pop-up' models from photos. |
| first_indexed | 2024-09-23T08:55:44Z |
| id | mit-1721.1/44615 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T08:55:44Z |
| publishDate | 2009 |
| record_format | dspace |
| spelling | mit-1721.1/446152019-04-10T19:48:02Z Automatic Class-Specific 3D Reconstruction from a Single Image Lozano-Perez, Tomas Kaelbling, Leslie Pack Chiu, Han-Pang Tomas Lozano-Perez Learning and Intelligent Systems Our goal is to automatically reconstruct 3D objects from a single image, by using prior 3D shape models of classes. The shape models, defined as a collection of oriented primitive shapes centered at fixed 3D positions, can be learned from a few labeled images for each class. The 3D class model can then be used to estimate the 3D shape of an object instance, including occluded parts, from a single image. We provide a quantitative evaluation of the shape estimation process on real objects and demonstrate its usefulness in three applications: robot manipulation, object detection, and generating 3D 'pop-up' models from photos. 2009-02-19T19:15:05Z 2009-02-19T19:15:05Z 2009-02-18 http://hdl.handle.net/1721.1/44615 MIT-CSAIL-TR-2009-008 9 p. application/pdf application/postscript |
| spellingShingle | Lozano-Perez, Tomas Kaelbling, Leslie Pack Chiu, Han-Pang Automatic Class-Specific 3D Reconstruction from a Single Image |
| title | Automatic Class-Specific 3D Reconstruction from a Single Image |
| title_full | Automatic Class-Specific 3D Reconstruction from a Single Image |
| title_fullStr | Automatic Class-Specific 3D Reconstruction from a Single Image |
| title_full_unstemmed | Automatic Class-Specific 3D Reconstruction from a Single Image |
| title_short | Automatic Class-Specific 3D Reconstruction from a Single Image |
| title_sort | automatic class specific 3d reconstruction from a single image |
| url | http://hdl.handle.net/1721.1/44615 |
| work_keys_str_mv | AT lozanopereztomas automaticclassspecific3dreconstructionfromasingleimage AT kaelblinglesliepack automaticclassspecific3dreconstructionfromasingleimage AT chiuhanpang automaticclassspecific3dreconstructionfromasingleimage |