scholarly journals Fully Automatic Facial Deformation Transfer

Symmetry ◽  
2019 ◽  
Vol 12 (1) ◽  
pp. 27 ◽  
Author(s):  
Shaojun Bian ◽  
Anzong Zheng ◽  
Lin Gao ◽  
Greg Maguire ◽  
Willem Kokke ◽  
...  
Keyword(s):  
3D Model ◽  
Detection Method ◽  
3D Models ◽  
Facial Landmark Detection ◽  
Facial Landmark ◽  
Complex Emotions ◽  
Deformation Transfer ◽  
Fully Automatic ◽  
Transfer Method ◽  
Mesh Models

Facial Animation is a serious and ongoing challenge for the Computer Graphic industry. Because diverse and complex emotions need to be expressed by different facial deformation and animation, copying facial deformations from existing character to another is widely needed in both industry and academia, to reduce time-consuming and repetitive manual work of modeling to create the 3D shape sequences for every new character. But transfer of realistic facial animations between two 3D models is limited and inconvenient for general use. Modern deformation transfer methods require correspondences mapping, in most cases, which are tedious to get. In this paper, we present a fast and automatic approach to transfer the deformations of the facial mesh models by obtaining the 3D point-wise correspondences in the automatic manner. The key idea is that we could estimate the correspondences with different facial meshes using the robust facial landmark detection method by projecting the 3D model to the 2D image. Experiments show that without any manual labelling efforts, our method detects reliable correspondences faster and simpler compared with the state-of-the-art automatic deformation transfer method on the facial models.

scholarly journals Detecting Facial Region and Landmarks at Once via Deep Network

Sensors ◽  
2021 ◽  
Vol 21 (16) ◽  
pp. 5360
Author(s):  
Taehyung Kim ◽  
Jiwon Mok ◽  
Euichul Lee
Keyword(s):  
Face Detection ◽  
Detection Method ◽  
Single Shot ◽  
Landmark Detection ◽  
Facial Landmark Detection ◽  
Facial Region ◽  
Facial Landmarks ◽  
Detection Model ◽  
Facial Landmark ◽  
Shot Detection

For accurate and fast detection of facial landmarks, we propose a new facial landmark detection method. Previous facial landmark detection models generally perform a face detection step before landmark detection. This greatly affects landmark detection performance depending on which face detection model is used. Therefore, we propose a model that can simultaneously detect a face region and a landmark without performing the face detection step before landmark detection. The proposed single-shot detection model is based on the framework of YOLOv3, a one-stage object detection method, and the loss function and structure are altered to learn faces and landmarks at the same time. In addition, EfficientNet-B0 was utilized as the backbone network to increase processing speed and accuracy. The learned database used 300W-LP with 64 facial landmarks. The average normalized error of the proposed model was 2.32 pixels. The processing time per frame was about 15 milliseconds, and the average precision of face detection was about 99%. As a result of the evaluation, it was confirmed that the single-shot detection model has better performance and speed than the previous methods. In addition, as a result of using the COFW database, which has 29 landmarks instead of 64 to verify the proposed method, the average normalization error was 2.56 pixels, which was also confirmed to show promising performance.

scholarly journals APPROACH TO 3D ANALYSIS OF GRAVITY PTOSIS

2017 ◽  
Vol XLII-2/W4 ◽  
pp. 123-127
Author(s):  
R. A. Guryanov ◽  
S. Monkin ◽  
A. Monkin ◽  
A. Petrov
Keyword(s):  
3D Model ◽  
3D Models ◽  
Operation Technique ◽  
3D Analysis ◽  
Visual Examination ◽  
Before And After ◽  
The Face ◽  
Ct Data ◽  
Mesh Models

The assessment of ptosis degree for rejuvenation procedures, the choice of following operation technique and evaluation of surgery result are based on subjective visual examination and surgeon’s experience.<br><br> The photogrammetric scans of 25 female patients of age 20 to 55 in vertical and supine (horizontal) position of body with placing the regular marker points on the face were analyzed. For 5 patients, also the CT data was acquired and segmentation of soft tissue was performed. Four of these patients underwent SMAS-lifting, the photogrammetry scanning was repeat 6 months after the operation.<br><br> Computer vision algorithms was used for markers detection on the 3D model texture, marker were projected from texture to triangular mesh. 3D mesh models were registered with user defined anatomy points and pair selection based on markers location was done. Pairs of points on vertical and horizontal 3D models were analyzed for surface tissue mobility examination.<br><br> The migration vectors of each side of the face are uniformly directed upwards and laterally. The vectors are projected at the areas of so-called ligaments demonstrate no evidence in deviation from row sequences.<br><br> The volume migration is strongly correlates with the age of examined patients, on the contrary the point migration moderately correlates with age in patients of 30 to 50 years old.<br><br> The analysis of migration vectors before and after the SMAS-lifting revealed no significant changes in surface points’ migration. The described method allows to assess the mechanical conditions of individual face and evaluate efficacy of surgery. This approach can be used for the classification of face ptosis grade.

Sign in / Sign up

Export Citation Format

Share Document