Statistical detection of independent movement from a moving camera
Least squares is perhaps the most commonly used method of parameter estimation in computer vision algorithms. However, the estimated parameters from a least squares fit can be corrupted beyond recognition in the presence of gross errors or outliers which plague any data from real imagery. Within thi...
| Main Authors: | , |
|---|---|
| Formato: | Journal article |
| Idioma: | English |
| Publicado: |
Elsevier
1993
|
| _version_ | 1826313894075826176 |
|---|---|
| author | Torr, PHS Murray, DW |
| author_facet | Torr, PHS Murray, DW |
| author_sort | Torr, PHS |
| collection | OXFORD |
| description | Least squares is perhaps the most commonly used method of parameter estimation in computer vision algorithms. However, the estimated parameters from a least squares fit can be corrupted beyond recognition in the presence of gross errors or outliers which plague any data from real imagery. Within this paper we present a general methodology to not only identify these outliers but also give indications about the reliability of a fit. The methods presented are then applied to the problem of motion segmentation, identifying the objects within an image moving independently of the background. The algorithm requires only the first order properties of the image intensities and does not require known camera motion. It has been tested on a variety of real imagery. A b-spline snake is initialized on the occluding contours of this region of interest. |
| first_indexed | 2024-09-25T04:23:34Z |
| format | Journal article |
| id | oxford-uuid:68199a3a-9c67-4bdb-a999-44a50e15c261 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-09-25T04:23:34Z |
| publishDate | 1993 |
| publisher | Elsevier |
| record_format | dspace |
| spelling | oxford-uuid:68199a3a-9c67-4bdb-a999-44a50e15c2612024-08-20T15:36:23ZStatistical detection of independent movement from a moving cameraJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:68199a3a-9c67-4bdb-a999-44a50e15c261EnglishSymplectic ElementsElsevier1993Torr, PHSMurray, DWLeast squares is perhaps the most commonly used method of parameter estimation in computer vision algorithms. However, the estimated parameters from a least squares fit can be corrupted beyond recognition in the presence of gross errors or outliers which plague any data from real imagery. Within this paper we present a general methodology to not only identify these outliers but also give indications about the reliability of a fit. The methods presented are then applied to the problem of motion segmentation, identifying the objects within an image moving independently of the background. The algorithm requires only the first order properties of the image intensities and does not require known camera motion. It has been tested on a variety of real imagery. A b-spline snake is initialized on the occluding contours of this region of interest. |
| spellingShingle | Torr, PHS Murray, DW Statistical detection of independent movement from a moving camera |
| title | Statistical detection of independent movement from a moving camera |
| title_full | Statistical detection of independent movement from a moving camera |
| title_fullStr | Statistical detection of independent movement from a moving camera |
| title_full_unstemmed | Statistical detection of independent movement from a moving camera |
| title_short | Statistical detection of independent movement from a moving camera |
| title_sort | statistical detection of independent movement from a moving camera |
| work_keys_str_mv | AT torrphs statisticaldetectionofindependentmovementfromamovingcamera AT murraydw statisticaldetectionofindependentmovementfromamovingcamera |