BabyNet: Reconstructing 3D faces of babies from uncalibrated photographs. (July 2023)

• Record Type: Journal Article
• Authors: Morales, Araceli; Alomar, Antonia; Porras, Antonio R.; Linguraru, Marius George; Piella, Gemma; Sukno, Federico M.
• Is Part Of: Pattern recognition. Volume 139(2023)
• Publication Date: 2023-07
• Subjects: 3D face reconstruction -- Graph neural network -- Baby model; 00-01 -- 99-00; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2023.109367
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: A specifically designed 3D face reconstruction system is needed to target babies. We present the BabyNet, which is constructed is constructed using graph neural networks. We use transfer learning to extract meaningful features from the input images. The BabyNet outperforms deep learning-based approached trained with adult data. The BabyNet outperforms classical model-fitting methods, even when baby-specific 3D morphable model (e.g. BabyFM) is used. Abstract: We present a 3D face reconstruction system that aims at recovering the 3D facial geometry of babies from uncalibrated photographs, BabyNet. Since the 3D facial geometry of babies differs substantially from that of adults, baby-specific facial reconstruction systems are needed. BabyNet consists of two stages: 1) a 3D graph convolutional autoencoder learns a latent space of the baby 3D facial shape; and 2) a 2D encoder that maps photographs to the 3D latent space based on representative features extracted using transfer learning. In this way, using the pre-trained 3D decoder, we can recover a 3D face from 2D images. We evaluate BabyNet and show that 1) methods based on adult datasets cannot model the 3D facial geometry of babies, which proves the need for a baby-specific method, and 2) BabyNet outperforms classical model-fitting methods even when a baby-specific 3D morphable model, such as BabyFM, is used.