Obj cut
In this paper we present a principled Bayesian method for detecting and segmenting instances of a particular object category within an image, providing a coherent methodology for combining top down and bottom up cues. The work draws together two powerful formulations: pictorial structures ( PS ) and...
| Main Authors: | , , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
IEEE Computer Society
2005
|
| _version_ | 1826315136683474944 |
|---|---|
| author | Kumar, MP Torr, PHS Zisserman, A |
| author_facet | Kumar, MP Torr, PHS Zisserman, A |
| author_sort | Kumar, MP |
| collection | OXFORD |
| description | In this paper we present a principled Bayesian method for detecting and segmenting instances of a particular object category within an image, providing a coherent methodology for combining top down and bottom up cues. The work draws together two powerful formulations: pictorial structures ( PS ) and Markov random fields (MRFs) both of which have efficient algorithms for their solution. The resulting combination, which we call the Object Category Specific MRF, suggests a solution to the problem that has long dogged MRFs namely that they provide a poor prior for specific shapes. In contrast, our model provides a prior that is global across the image plane using the PS. We develop an efficient method, OBJ CUT, to obtain segmentations using this model. Novel aspects of this method include an efficient algorithm for sampling the PS model, and the observation that the expected log likelihood of the model can be increased by a single graph cut. Results are presented on two object categories, cows and horses. We compare our methods to the state of the art in object category specific image segmentation and demonstrate significant improvements. |
| first_indexed | 2024-12-09T03:20:23Z |
| format | Conference item |
| id | oxford-uuid:7e37e1cc-6d18-4326-8032-96c452a98cc9 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-12-09T03:20:23Z |
| publishDate | 2005 |
| publisher | IEEE Computer Society |
| record_format | dspace |
| spelling | oxford-uuid:7e37e1cc-6d18-4326-8032-96c452a98cc92024-11-05T13:14:47ZObj cutConference itemhttp://purl.org/coar/resource_type/c_5794uuid:7e37e1cc-6d18-4326-8032-96c452a98cc9EnglishSymplectic ElementsIEEE Computer Society2005Kumar, MPTorr, PHSZisserman, AIn this paper we present a principled Bayesian method for detecting and segmenting instances of a particular object category within an image, providing a coherent methodology for combining top down and bottom up cues. The work draws together two powerful formulations: pictorial structures ( PS ) and Markov random fields (MRFs) both of which have efficient algorithms for their solution. The resulting combination, which we call the Object Category Specific MRF, suggests a solution to the problem that has long dogged MRFs namely that they provide a poor prior for specific shapes. In contrast, our model provides a prior that is global across the image plane using the PS. We develop an efficient method, OBJ CUT, to obtain segmentations using this model. Novel aspects of this method include an efficient algorithm for sampling the PS model, and the observation that the expected log likelihood of the model can be increased by a single graph cut. Results are presented on two object categories, cows and horses. We compare our methods to the state of the art in object category specific image segmentation and demonstrate significant improvements. |
| spellingShingle | Kumar, MP Torr, PHS Zisserman, A Obj cut |
| title | Obj cut |
| title_full | Obj cut |
| title_fullStr | Obj cut |
| title_full_unstemmed | Obj cut |
| title_short | Obj cut |
| title_sort | obj cut |
| work_keys_str_mv | AT kumarmp objcut AT torrphs objcut AT zissermana objcut |