A Method for the Automatic Normalization and 3D Facial Mesh Extraction from Head Magnetic Resonance Imaging
Sevillano, X.; Urena, D.; Gonzalez, R.; Fatjo-Vilas, M.; Salvador, R.; Pomarol-Clotet, E.; Martinez-Abadias, N.
Show abstract
The analysis of 3D facial shape in medicine is motivated by the fact that certain diseases and syndromes are associated to specific facial dysmorphologies. In this context, 3D facial shape analysis constitutes a promising and non-invasive support to traditional diagnostic methods. In this work, we explore the use of head magnetic resonances to obtain accurate 3D facial meshes that enable subsequent facial shape analysis. We present a fully automatic method that normalizes the orientation and alignment of 3D point clouds corresponding to head magnetic resonances by detecting salient facial features. Moreover, using clustering techniques, our method also allows to eliminate noise and artifacts appearing in magnetic resonance imaging. Finally, through bidirectional ray tracing, we obtain a dense 3D facial mesh that accurately captures facial shape. The proposed method has been built and evaluated on a dataset of 185 head magnetic resonances, and it has demonstrated its ability to successfully orient, align and obtain a dense 3D facial mesh with a high accuracy rate.
Matching journals
The top 7 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Classification and Visualisation of Normal and Abnormal Radiographs; a comparison between Eleven Convolutional Neural Network Architectures 93%
- High-Throughput and Accurate 3D Scanning of Cattle Using Time-of-Flight Sensors and Deep Learning 91%
- Scorepochs: a computer-aided scoring tool for resting-state M/EEG epochs 91%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.