3D Face Reconstruction by Learning from Synthetic Data
Gespeichert in:
| Titel: | 3D Face Reconstruction by Learning from Synthetic Data |
|---|---|
| Autoren: | Richardson, Elad, Sela, Matan, Kimmel, Ron |
| Quelle: | 2016 Fourth International Conference on 3D Vision (3DV) |
| Publication Status: | Preprint |
| Verlagsinformationen: | IEEE, 2016. |
| Publikationsjahr: | 2016 |
| Schlagwörter: | FOS: Computer and information sciences, Computer Vision and Pattern Recognition (cs.CV), Computer Science - Computer Vision and Pattern Recognition, 0202 electrical engineering, electronic engineering, information engineering, 02 engineering and technology |
| Beschreibung: | Fast and robust three-dimensional reconstruction of facial geometric structure from a single image is a challenging task with numerous applications. Here, we introduce a learning-based approach for reconstructing a three-dimensional face from a single image. Recent face recovery methods rely on accurate localization of key characteristic points. In contrast, the proposed approach is based on a Convolutional-Neural-Network (CNN) which extracts the face geometry directly from its image. Although such deep architectures outperform other models in complex computer vision problems, training them properly requires a large dataset of annotated examples. In the case of three-dimensional faces, currently, there are no large volume data sets, while acquiring such big-data is a tedious task. As an alternative, we propose to generate random, yet nearly photo-realistic, facial images for which the geometric form is known. The suggested model successfully recovers facial shapes from real images, even for faces with extreme expressions and under various lighting conditions. The first two authors contributed equally to this work |
| Publikationsart: | Article Conference object |
| DOI: | 10.1109/3dv.2016.56 |
| DOI: | 10.48550/arxiv.1609.04387 |
| Zugangs-URL: | http://arxiv.org/pdf/1609.04387 http://arxiv.org/abs/1609.04387 https://arxiv.org/pdf/1609.04387 https://dblp.uni-trier.de/db/journals/corr/corr1609.html#RichardsonSK16 https://ui.adsabs.harvard.edu/abs/2016arXiv160904387R/abstract https://ieeexplore.ieee.org/abstract/document/7785121 https://arxiv.org/abs/1609.04387 https://doi.org/10.1109/3DV.2016.56 |
| Rights: | arXiv Non-Exclusive Distribution |
| Dokumentencode: | edsair.doi.dedup.....75f1e406318bc114eefa840ad1c6e0f1 |
| Datenbank: | OpenAIRE |
Nájsť tento článok vo Web of Science | Abstract: | Fast and robust three-dimensional reconstruction of facial geometric structure from a single image is a challenging task with numerous applications. Here, we introduce a learning-based approach for reconstructing a three-dimensional face from a single image. Recent face recovery methods rely on accurate localization of key characteristic points. In contrast, the proposed approach is based on a Convolutional-Neural-Network (CNN) which extracts the face geometry directly from its image. Although such deep architectures outperform other models in complex computer vision problems, training them properly requires a large dataset of annotated examples. In the case of three-dimensional faces, currently, there are no large volume data sets, while acquiring such big-data is a tedious task. As an alternative, we propose to generate random, yet nearly photo-realistic, facial images for which the geometric form is known. The suggested model successfully recovers facial shapes from real images, even for faces with extreme expressions and under various lighting conditions.<br />The first two authors contributed equally to this work |
|---|---|
| DOI: | 10.1109/3dv.2016.56 |