|
|
Stereoscopic Face Images Matching
Klaudíny, Martin ; Kršek, Přemysl (referee) ; Zemčík, Pavel (advisor)
This thesis is dedicated to the problem of 3D face capture. A majority of current face capture systems exploits a pattern projection. The goal of thesis was to develop the 3D face capture system based on the passive stereo photogrammetry. The high-resolution images of a face are captured by one pair of the calibrated digital still cameras. The image pre-processing consists of the removal of lens distortion, rectification of image pair and face extraction. The stereo image matching is accentuated. Three different approaches to the stereo correspondence search are presented - the local technique, the global technique using a graph cut and the hybrid technique merging previous two. At last, the 3D model is reconstructed according to the found correspondence. The results show that the high-quality 3D face models can be obtained despite traditional difficulties with matching pattern-free face images. Also it is possible to keep the computational demands on the acceptable level, although the high-resolution images are processed.
|
|
|
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.
|
|
|
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.
|
|
|
Stereoscopic Face Images Matching
Klaudíny, Martin ; Kršek, Přemysl (referee) ; Zemčík, Pavel (advisor)
This thesis is dedicated to the problem of 3D face capture. A majority of current face capture systems exploits a pattern projection. The goal of thesis was to develop the 3D face capture system based on the passive stereo photogrammetry. The high-resolution images of a face are captured by one pair of the calibrated digital still cameras. The image pre-processing consists of the removal of lens distortion, rectification of image pair and face extraction. The stereo image matching is accentuated. Three different approaches to the stereo correspondence search are presented - the local technique, the global technique using a graph cut and the hybrid technique merging previous two. At last, the 3D model is reconstructed according to the found correspondence. The results show that the high-quality 3D face models can be obtained despite traditional difficulties with matching pattern-free face images. Also it is possible to keep the computational demands on the acceptable level, although the high-resolution images are processed.
|
guest ::
