Gaussian process latent variable models for human pose estimation
We describe a method for recovering 3D human body pose from silhouettes. Our model is based on learning a latent space using the Gaussian Process Latent Variable Model (GP-LVM) [1] encapsulating both pose and silhouette features Our method is generative, this allows us to model the ambiguities of a...
| 主要な著者: | , , |
|---|---|
| フォーマット: | Conference item |
| 言語: | English |
| 出版事項: |
Springer
2008
|
| _version_ | 1826315245626327040 |
|---|---|
| author | Ek, CH Torr, PHS Lawrence, ND |
| author_facet | Ek, CH Torr, PHS Lawrence, ND |
| author_sort | Ek, CH |
| collection | OXFORD |
| description | We describe a method for recovering 3D human body pose from silhouettes. Our model is based on learning a latent space using the Gaussian Process Latent Variable Model (GP-LVM) [1] encapsulating both pose and silhouette features Our method is generative, this allows us to model the ambiguities of a silhouette representation in a principled way. We learn a dynamical model over the latent space which allows us to disambiguate between ambiguous silhouettes by temporal consistency. The model has only two free parameters and has several advantages over both regression approaches and other generative methods. In addition to the application shown in this paper the suggested model is easily extended to multiple observation spaces without constraints on type. |
| first_indexed | 2024-12-09T03:22:21Z |
| format | Conference item |
| id | oxford-uuid:7c19e12d-e1eb-49b4-90f0-d850a1fb6a98 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-12-09T03:22:21Z |
| publishDate | 2008 |
| publisher | Springer |
| record_format | dspace |
| spelling | oxford-uuid:7c19e12d-e1eb-49b4-90f0-d850a1fb6a982024-11-18T11:06:39ZGaussian process latent variable models for human pose estimationConference itemhttp://purl.org/coar/resource_type/c_5794uuid:7c19e12d-e1eb-49b4-90f0-d850a1fb6a98EnglishSymplectic ElementsSpringer2008Ek, CHTorr, PHSLawrence, NDWe describe a method for recovering 3D human body pose from silhouettes. Our model is based on learning a latent space using the Gaussian Process Latent Variable Model (GP-LVM) [1] encapsulating both pose and silhouette features Our method is generative, this allows us to model the ambiguities of a silhouette representation in a principled way. We learn a dynamical model over the latent space which allows us to disambiguate between ambiguous silhouettes by temporal consistency. The model has only two free parameters and has several advantages over both regression approaches and other generative methods. In addition to the application shown in this paper the suggested model is easily extended to multiple observation spaces without constraints on type. |
| spellingShingle | Ek, CH Torr, PHS Lawrence, ND Gaussian process latent variable models for human pose estimation |
| title | Gaussian process latent variable models for human pose estimation |
| title_full | Gaussian process latent variable models for human pose estimation |
| title_fullStr | Gaussian process latent variable models for human pose estimation |
| title_full_unstemmed | Gaussian process latent variable models for human pose estimation |
| title_short | Gaussian process latent variable models for human pose estimation |
| title_sort | gaussian process latent variable models for human pose estimation |
| work_keys_str_mv | AT ekch gaussianprocesslatentvariablemodelsforhumanposeestimation AT torrphs gaussianprocesslatentvariablemodelsforhumanposeestimation AT lawrencend gaussianprocesslatentvariablemodelsforhumanposeestimation |