Automatic Voice Synthesis System using Long Short-Term Memory and Google Mediapipe

Varghese, Ria; Cherian, Reena Mary; Ninan, Sanjana R.; Arun, Nikita; Anjusree, V.K

doi:10.1109/icccnt51525.2021.9579837

Cited by 1 publication

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed approach was used to locate and extract features automatically. In the present work, 468 landmarks were extracted instead of 68 to improve the accuracy using the Mediapipe library [36], [37], [38], [39], [40], [41]. The media pipe framework was used to detect multiple faces and 468 face landmarks in a 3D space.…”

Section: Research Methods a Methodologymentioning

confidence: 99%

“…• Mediapipe Framework [36], [37], [38], [39], [40], [41]: is an open-source, cross-platform, face geometry solution library developed by Google for computer vision tasks. It estimated 468, 3D face landmarks in realtime, even on mobile devices.…”

Section: ) Comparison Of Established Methodsmentioning

confidence: 99%

“…In the proposed approach, after detecting the face position in an image, the next step is to locate 468 landmarks using a media pipe [30], [31], [32], [33], [34]. Mediapipe is an open-source framework for ''building world-class machine learning solutions'' using Google which is fast and highly accurate [31], [35], [36], [37], [38], [39], [40], [41]. 2D facial images are used to detect 3D face meshes (which include [60], [61], [62], [63] of the self-created dataset and face mesh generation from the extracted landmarks.…”

Section: ) Digital Facial Anthropometrymentioning

confidence: 99%

“…Face landmark extraction must be quick and accurate to meet the demands of various capabilities, such as real-time processing or mobile device rendering [26], [27], [28], [29]. Precise recognition of landmarks is performed using Mediapipe [31], [32], [33], [34], [35] which is mainly used in real-time applications such as emotion detection, Parkinson's disease detection, driver drowsiness detection, and earlystage autism screening [36], [37], [38], [39], [40], [41]. It estimates 468 landmarks in real-time to improve the accuracy of the face recognition system (FRS) compared to other existing approaches, such as Multi-Task Cascaded Convolutional Networks (MTCNN) [42], [43] and Digital Library (DLIB) [44], [45].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

2D-3D Facial Image Analysis for Identification of Facial Features Using Machine Learning Algorithms With Hyper-Parameter Optimization for Forensics Applications

et al. 2023

View full text Add to dashboard Cite

Recognizing a face is a remarkable process that humans naturally use. Computer vision has tried to resemble this ability of human vision as a biometric tool to identify humans. Commercial and law enforcement applications are increasingly using face recognition technology to identify people. Currently, it is one of the most sought-after detection methods used in forensics for criminal identification purposes. Owing to similarities in the appearance of certain faces, especially in criminal cases, this problem poses a great challenge in forensic investigation and detection. The novelty of this work lies in the development of a framework for face recognition using 2D facial images gathered from various sources to generate a 3D face mesh using 468 MediaPipe landmarks which detects multiple faces in real-time. This leads to the generation of input feature vectors being formulated utilizing Euclidean/Geodesic distances and their ratios between the landmarks. These feature vectors are then trained into various classifiers that can provide the correct matching decision in an unrestricted environment such as large pose, expression, and occlusion variations. These quantitative similarity measures can then be presented as statistical evidence to identify criminals in forensic investigations. This two-dimensional to three-dimensional annotation provides a higher quality of threedimensional reconstructed face models without the need for any additional three-dimensional morphable models. The proposed methods were validated and tested to achieve comparable recognition performance using hyperparameter optimization. Regarding accuracy, the statistical results show that Extreme gradient boosting is the best classification model that provides the best accuracy (78%) for predicting facial images compared with Adaptive Boosting (77%), Random Forest (75%), Bernoulli Naive Bayes (68%), Decision Tree (57%), Logistic Regression (71%), Light Gradient Boosting Model (58%), Extra Tree Classifier (57%), Support Vector Machine (58%), and Nearest Centroid (62%) classifiers which can be further increased by considering a greater number of images in the dataset implying at the potential of further research for scale-up implementation.

show abstract