Unsupervised Disentanglement of Pose, Appearance and Background from Images and Videos.
IEEE Trans Pattern Anal Mach Intell
; 44(7): 3883-3894, 2022 Jul.
Article
en En
| MEDLINE
| ID: mdl-33513098
ABSTRACT
Unsupervised landmark learning is the task of learning semantic keypoint-like representations without the use of expensive input keypoint annotations. A popular approach is to factorize an image into a pose and appearance data stream, then to reconstruct the image from the factorized components. The pose representation should capture a set of consistent and tightly localized landmarks in order to facilitate reconstruction of the input image. Ultimately, we wish for our learned landmarks to focus on the foreground object of interest. However, the reconstruction task of the entire image forces the model to allocate landmarks to model the background. Using a motion-based foreground assumption, this work explores the effects of factorizing the reconstruction task into separate foreground and background reconstructions in an unsupervised way, allowing the model to condition only the foreground reconstruction on the unsupervised landmarks. Our experiments demonstrate that the proposed factorization results in landmarks that are focused on the foreground object of interest when measured against ground-truth foreground masks. Furthermore, the rendered background quality is also improved as ill-suited landmarks are no longer forced to model this content. We demonstrate this improvement via improved image fidelity in a video-prediction task. Code is available at https//github.com/NVIDIA/UnsupervisedLandmarkLearning.
Texto completo:
1
Colección:
01-internacional
Banco de datos:
MEDLINE
Tipo de estudio:
Prognostic_studies
Idioma:
En
Revista:
IEEE Trans Pattern Anal Mach Intell
Asunto de la revista:
INFORMATICA MEDICA
Año:
2022
Tipo del documento:
Article