Unsupervised learning of probably symmetric deformable 3D objects from images in the wild
We propose a method to learn 3D deformable object categories from raw single-view images, without external supervision. The method is based on an autoencoder that factors each input image into depth, albedo, viewpoint and illumination. In order to disentangle these components without supervision, we...
| 主要な著者: | , , |
|---|---|
| フォーマット: | Journal article |
| 言語: | English |
| 出版事項: |
IEEE
2021
|
| _version_ | 1826309867111972864 |
|---|---|
| author | Wu, S Rupprecht, C Vedaldi, A |
| author_facet | Wu, S Rupprecht, C Vedaldi, A |
| author_sort | Wu, S |
| collection | OXFORD |
| description | We propose a method to learn 3D deformable object categories from raw single-view images, without external supervision. The method is based on an autoencoder that factors each input image into depth, albedo, viewpoint and illumination. In order to disentangle these components without supervision, we use the fact that many object categories have, at least approximately, a symmetric structure. We show that reasoning about illumination allows us to exploit the underlying object symmetry even if the appearance is not symmetric due to shading. Furthermore, we model objects that are probably, but not certainly, symmetric by predicting a symmetry probability map, learned end-to-end with the other components of the model. Our experiments show that this method can recover very accurately the 3D shape of human faces, cat faces and cars from single-view images, without any supervision or a prior shape model. On benchmarks, we demonstrate superior accuracy compared to another method that uses supervision at the level of 2D image correspondences.
|
| first_indexed | 2024-03-07T07:42:07Z |
| format | Journal article |
| id | oxford-uuid:344a210a-c921-4c05-befd-2c138837b5db |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T07:42:07Z |
| publishDate | 2021 |
| publisher | IEEE |
| record_format | dspace |
| spelling | oxford-uuid:344a210a-c921-4c05-befd-2c138837b5db2023-05-02T12:25:20ZUnsupervised learning of probably symmetric deformable 3D objects from images in the wildJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:344a210a-c921-4c05-befd-2c138837b5dbEnglishSymplectic ElementsIEEE2021Wu, SRupprecht, CVedaldi, AWe propose a method to learn 3D deformable object categories from raw single-view images, without external supervision. The method is based on an autoencoder that factors each input image into depth, albedo, viewpoint and illumination. In order to disentangle these components without supervision, we use the fact that many object categories have, at least approximately, a symmetric structure. We show that reasoning about illumination allows us to exploit the underlying object symmetry even if the appearance is not symmetric due to shading. Furthermore, we model objects that are probably, but not certainly, symmetric by predicting a symmetry probability map, learned end-to-end with the other components of the model. Our experiments show that this method can recover very accurately the 3D shape of human faces, cat faces and cars from single-view images, without any supervision or a prior shape model. On benchmarks, we demonstrate superior accuracy compared to another method that uses supervision at the level of 2D image correspondences. |
| spellingShingle | Wu, S Rupprecht, C Vedaldi, A Unsupervised learning of probably symmetric deformable 3D objects from images in the wild |
| title | Unsupervised learning of probably symmetric deformable 3D objects from images in the wild |
| title_full | Unsupervised learning of probably symmetric deformable 3D objects from images in the wild |
| title_fullStr | Unsupervised learning of probably symmetric deformable 3D objects from images in the wild |
| title_full_unstemmed | Unsupervised learning of probably symmetric deformable 3D objects from images in the wild |
| title_short | Unsupervised learning of probably symmetric deformable 3D objects from images in the wild |
| title_sort | unsupervised learning of probably symmetric deformable 3d objects from images in the wild |
| work_keys_str_mv | AT wus unsupervisedlearningofprobablysymmetricdeformable3dobjectsfromimagesinthewild AT rupprechtc unsupervisedlearningofprobablysymmetricdeformable3dobjectsfromimagesinthewild AT vedaldia unsupervisedlearningofprobablysymmetricdeformable3dobjectsfromimagesinthewild |