|
|
Learning the Face Behind a Voice
Kyjonka, Mojmír ; Matějka, Pavel (referee) ; Plchot, Oldřich (advisor)
This thesis deals with face reconstruction based on voice. The state of the art of this problem is investigated and model for such problem is trained. Model used in this thesis is based on the work "Reconstructing faces from voices" which architecture is based on Generative Adversarial Network (GAN). In this work, we used VGGFace and VoxCeleb datasets, and additionally, we created a small audiovisual dataset of Czech speakers. This work was implemented using the Python scripting language and PyTorch library.
|
|
|
Creating a 3D Face Model Using Kinect Device
Nesvadba, Tomáš ; Malinka, Kamil (referee) ; Mráček, Štěpán (advisor)
The main aim of this bachelor's thesis is to suggest a method for increasing the resolution of 3D face models captured by Kinect. The document describes different methods of capturing 3D models and compares the Kinect sensor with more acurate scanners. The main part of this document is dedicated to statistic model based on the PCA and its usage in the project. The last section of the paper describes testing of various models and discuss the results and another options of aplication.
|
|
|
3D Face Reconstruction Based on 2D Image
Karhánek, Martin ; Láník, Aleš (referee) ; Částek, Petr (advisor)
This work deals with procedures that enables you reconstruct 3D face from a 2D picture. It describes ways to analyze an input picture, such as a face and facial features localization, on which this reconstruction builds. A reader is getting into a morphable model, its creation from static database data and and its usage on a reconstruction, more detailed later on, this morphable model is a building block of mentioned methods. This work contains also an algorithm implementation based on this model.
|
|
|
From the facial reconstruction based on the skull to the identification of the individual: demands, principles, problems
Moštková, Miroslava ; Brůžek, Jaroslav (advisor) ; Blažek, Vladimír (referee)
Facial reconstruction based on the skull is a technique allowing recreation of the original facial features of an individual. However, recreation of the exact look is nearly impossible at this time. Facial reconstruction is used during archaeological research or during investigation of forensic cases. The facial reconstruction methods used are morphoscopic, morphometric, or a combination of the two. They are used during a manual and computerized process to create two dimensional or three-dimensional reconstruction of the individual. Accuracy and reliability can be determined with each one of these methods by quantitative and qualitative measurements. Reliability and accuracy of the facial reconstruction should also be considered from a view of the facial perception. The human face is perceived in a holistic-analytical way which is based on a calculation of the distances between different features. Facial reconstruction is based on the recognition of familiar faces that are perceived on the basis of internal features that are not influenced by differences in the view angle or expression. Unfamiliar faces are perceived on the basis of from external features. Faces with significant facial features are identified faster. Facial features can be organized by the significance of their influence for recognition of...
|
|
|
Facial reconstruction: an error of facial features prediction and facial recognition
Trsová, Iva ; Zedníková Malá, Pavla (advisor) ; Eliášová, Hana (referee)
Approximate facial reconstruction by features of the skull is one of the ways of estimating a person's appearance and assessing their identity. However, its use in practice has often been criticised. Even in our modern age, it is not an exclusively scientific method with an outstanding success ratio that would under any circumstances lead to identification of the target individual. It is applied primarily in archeology and forensic anthropology. There are more predictions methods for each part of the face. Each author prefers a different one and new ones are still emerging. The aim of the thesis is to provide an overview of the current prediction methods used in frontal facial reconstruction and also provide information about their precision/reliability. Furthermore, the thesis introduces the matter of facial reconstruction with focus on the changes in the position or size of the facial features (eyes, nose, lips). Key words: facial reconstruction, facial prediction, methods of reconstruction, change the position of facial features, accuracy
|
|
|
Learning the Face Behind a Voice
Kyjonka, Mojmír ; Matějka, Pavel (referee) ; Plchot, Oldřich (advisor)
This thesis deals with face reconstruction based on voice. The state of the art of this problem is investigated and model for such problem is trained. Model used in this thesis is based on the work "Reconstructing faces from voices" which architecture is based on Generative Adversarial Network (GAN). In this work, we used VGGFace and VoxCeleb datasets, and additionally, we created a small audiovisual dataset of Czech speakers. This work was implemented using the Python scripting language and PyTorch library.
|
|
|
Facial reconstruction: an error of facial features prediction and facial recognition
Trsová, Iva ; Zedníková Malá, Pavla (advisor) ; Bejdová, Šárka (referee)
Approximate facial reconstruction by features of the skull is one of the ways of estimating a person's appearance and assessing their identity. However, its use in practice has often been criticised. Even in our modern age, it is not an exclusively scientific method with an outstanding success ratio that would under any circumstances lead to identification of the target individual. It is applied primarily in archeology and forensic anthropology. There are more predictions methods for each part of the face. Each author prefers a different one and new ones are still emerging. The aim of the thesis is to provide an overview of the current prediction methods used in frontal facial reconstruction and also provide information about their precision/reliability. Furthermore, the thesis introduces the matter of facial recognition with focus on the influence of the changes in the position or size of the facial features (eyes, nose, lips). Key words: facial reconstruction, facial prediction, methods of reconstruction, face recognition, change of the position of facial features, accuracy of prediction guides
|
|
|
Facial reconstruction: an error of facial features prediction and facial recognition
Trsová, Iva ; Zedníková Malá, Pavla (advisor) ; Eliášová, Hana (referee)
Approximate facial reconstruction by features of the skull is one of the ways of estimating a person's appearance and assessing their identity. However, its use in practice has often been criticised. Even in our modern age, it is not an exclusively scientific method with an outstanding success ratio that would under any circumstances lead to identification of the target individual. It is applied primarily in archeology and forensic anthropology. There are more predictions methods for each part of the face. Each author prefers a different one and new ones are still emerging. The aim of the thesis is to provide an overview of the current prediction methods used in frontal facial reconstruction and also provide information about their precision/reliability. Furthermore, the thesis introduces the matter of facial reconstruction with focus on the changes in the position or size of the facial features (eyes, nose, lips). Key words: facial reconstruction, facial prediction, methods of reconstruction, change the position of facial features, accuracy
|
|
|
From the facial reconstruction based on the skull to the identification of the individual: demands, principles, problems
Moštková, Miroslava ; Brůžek, Jaroslav (advisor) ; Blažek, Vladimír (referee)
Facial reconstruction based on the skull is a technique allowing recreation of the original facial features of an individual. However, recreation of the exact look is nearly impossible at this time. Facial reconstruction is used during archaeological research or during investigation of forensic cases. The facial reconstruction methods used are morphoscopic, morphometric, or a combination of the two. They are used during a manual and computerized process to create two dimensional or three-dimensional reconstruction of the individual. Accuracy and reliability can be determined with each one of these methods by quantitative and qualitative measurements. Reliability and accuracy of the facial reconstruction should also be considered from a view of the facial perception. The human face is perceived in a holistic-analytical way which is based on a calculation of the distances between different features. Facial reconstruction is based on the recognition of familiar faces that are perceived on the basis of internal features that are not influenced by differences in the view angle or expression. Unfamiliar faces are perceived on the basis of from external features. Faces with significant facial features are identified faster. Facial features can be organized by the significance of their influence for recognition of...
|
|
|
Creating a 3D Face Model Using Kinect Device
Nesvadba, Tomáš ; Malinka, Kamil (referee) ; Mráček, Štěpán (advisor)
The main aim of this bachelor's thesis is to suggest a method for increasing the resolution of 3D face models captured by Kinect. The document describes different methods of capturing 3D models and compares the Kinect sensor with more acurate scanners. The main part of this document is dedicated to statistic model based on the PCA and its usage in the project. The last section of the paper describes testing of various models and discuss the results and another options of aplication.
|
guest ::
