• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 6
  • 1
  • 1
  • 1
  • Tagged with
  • 11
  • 11
  • 7
  • 7
  • 5
  • 3
  • 3
  • 3
  • 3
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Towards the Development of Training Tools for Face Recognition

Rodriguez, Jobany 2011 May 1900 (has links)
Distinctiveness plays an important role in the recognition of faces, i.e., a distinctive face is usually easier to remember than a typical face in a recognition task. This distinctiveness effect explains why caricatures are recognized faster and more accurately than unexaggerated (i.e., veridical) faces. Furthermore, using caricatures during training can facilitate recognition of a person’s face at a later time. The objective of this thesis is to determine the extent to which photorealistic computer-generated caricatures may be used in training tools to improve recognition of faces by humans. To pursue this objective, we developed a caricaturization procedure for three-dimensional (3D) face models, and characterized face recognition performance (by humans) through a series of perceptual studies. The first study focused on 3D shape information without texture. Namely, we tested whether exposure to caricatures during an initial familiarization phase would aid in the recognition of their veridical counterparts at a later time. We examined whether this effect would emerge with frontal rather than three-quarter views, after very brief exposure to caricatures during the learning phase and after modest rotations of faces during the recognition phase. Results indicate that, even under these difficult training conditions, people are more accurate at recognizing unaltered faces if they are first familiarized with caricatures of the faces, rather than with the unaltered faces. These preliminary findings support the use of caricatures in new training methods to improve face recognition. In the second study, we incorporated texture into our 3D models, which allowed us to generate photorealistic renderings. In this study, we sought to determine the extent to which familiarization with caricaturized faces could also be used to reduce other-race effects (e.g., the phenomenon whereby faces from other races appear less distinct than faces from our own race). Using an old/new face recognition paradigm, Caucasian participants were first familiarized with a set of faces from multiple races, and then asked to recognize those faces among a set of confounders. Participants who were familiarized with and then asked to recognize veridical versions of the faces showed a significant other-race effect on Indian faces. In contrast, participants who were familiarized with caricaturized versions of the same faces, and then asked to recognize their veridical versions, showed no other-race effects on Indian faces. This result suggests that caricaturization may be used to help individuals focus their attention to features that are useful for recognition of other-race faces. The third and final experiment investigated the practical application of our earlier results. Since 3D facial scans are not generally available, here we also sought to determine whether 3D reconstructions from 2D frontal images could be used for the same purpose. Using the same old/new face recognition paradigm, participants who were familiarized with reconstructed faces and then asked to recognize the ground truth versions of the faces showed a significant reduction in performance compared to the previous study. In addition, participants who were familiarized with caricatures of reconstructed versions, and then asked to recognize their corresponding ground truth versions, showed a larger reduction in performance. Our results suggest that, despite the high level of photographic realism achieved by current 3D facial reconstruction methods, additional research is needed in order to reduce reconstruction errors and capture the distinctive facial traits of an individual. These results are critical for the development of training tools based on computer-generated photorealistic caricatures from “mug shot” images.
2

3d Face Reconstruction Using Stereo Vision

Dikmen, Mehmet 01 September 2006 (has links) (PDF)
3D face modeling is currently a popular area in Computer Graphics and Computer Vision. Many techniques have been introduced for this purpose, such as using one or more cameras, 3D scanners, and many other systems of sophisticated hardware with related software. But the main goal is to find a good balance between visual reality and the cost of the system. In this thesis, reconstruction of a 3D human face from a pair of stereo cameras is studied. Unlike many other systems, facial feature points are obtained automatically from two photographs with the help of a dot pattern projected on the object&amp / #8217 / s face. It is seen that using projection pattern also provided enough feature points to derive 3D face roughly. These points are then used to fit a generic face mesh for a more realistic model. To cover this 3D model, a single texture image is generated from the initial stereo photographs.
3

Optimizing Realistic 3D Facial Models for VR Avatars through Mesh Simplification / Optimering av realistiska 3D-ansiktsmodeller för VR-avatarer genom nätverksförenkling

Liu, Beiqian January 2023 (has links)
The use of realistic 3D avatars in Virtual Reality (VR) has gained significant traction in applications such as telecommunication and gaming, offering immersive experiences and face-to-face interactions. However, standalone VR devices often face limitations in computational resources and real-time rendering requirements, necessitating the optimization of 3D models through mesh simplification to enhance performance and ensure a smooth user experience. This thesis presents a pipeline that utilizes a Convolutional Neural Network to reconstruct realistic 3D human facial models in a static form from single RGB head images. The reconstructed models are then subjected to the Quadric Error Metrics simplification algorithm, enabling different levels of simplification to be achieved. An evaluation was conducted, utilizing 30 photos from the NoW dataset, to examine the trade-offs associated with employing mesh simplification on the generated facial models within the VR environment. The evaluation results demonstrate that reducing the polygon count improves frame rates and reduces GPU usage in VR, thereby enhancing overall performance. However, this improvement comes at the cost of increased simplification execution time and geometric errors, and decreased perceptual quality. This research contributes to the understanding of mesh simplification’s impact on human facial models within the VR context, providing insights into balancing model complexity and real-time rendering performance, particularly in resource-constrained environments such as mobile devices or cloud-based applications, as well as for models located farther away from the cameras. / Användningen av realistiska 3D-avatarer inom Virtuell Verklighet (VR) har fått betydande uppmärksamhet inom tillämpningar som telekommunikation och spel, vilket erbjuder en uppslukande upplevelse och ansikte mot ansikte-interaktioner. Dock möter fristående VR-enheter ofta begränsningar i beräkningsresurser och krav på realtidsrendering, vilket gör det nödvändigt att optimera 3D-modeller genom nätverksförenkling för att förbättra prestanda och säkerställa en smidig användarupplevelse. Denna avhandling presenterar en pipeline som använder sig av ett konvolutionellt neuralt nätverk för att rekonstruera realistiska 3D-modeller av mänskliga ansikten i en statisk form från enstaka RGB-bilder av huvudet. De rekonstruerade modellerna genomgår sedan nätverksförenkling med Quadric Error Metrics-algoritmen, vilket möjliggör olika nivåer av förenkling. En utvärdering genomfördes, med hjälp av 30 foton från NoW-datasetet, för att undersöka avvägningarna i samband med att använda nätverksförenkling på de genererade ansiktsmodellerna inom VR-miljön. Utvärderingsresultaten visar att minskning av polygonantal förbättrar bildhastigheten och minskar GPU-användningen inom VR, vilket förbättrar den övergripande prestandan. Dock sker denna förbättring på bekostnad av ökad tid för förenklingsexekvering och geometriska fel, samt minskad perceptuell kvalitet. Denna forskning bidrar till förståelsen av nätverksförenklingens påverkan på mänskliga ansiktsmodeller inom VR-sammanhanget och ger insikter om att balansera modellkomplexitet och realtidsrenderingsprestanda, särskilt i resursbegränsade miljöer som mobilenheter eller molnbaserade applikationer, samt för modeller som är längre bort från kamerorna.
4

3D Human Face Reconstruction and 2D Appearance Synthesis

Zhao, Yajie 01 January 2018 (has links)
3D human face reconstruction has been an extensive research for decades due to its wide applications, such as animation, recognition and 3D-driven appearance synthesis. Although commodity depth sensors are widely available in recent years, image based face reconstruction are significantly valuable as images are much easier to access and store. In this dissertation, we first propose three image-based face reconstruction approaches according to different assumption of inputs. In the first approach, face geometry is extracted from multiple key frames of a video sequence with different head poses. The camera should be calibrated under this assumption. As the first approach is limited to videos, we propose the second approach then focus on single image. This approach also improves the geometry by adding fine grains using shading cue. We proposed a novel albedo estimation and linear optimization algorithm in this approach. In the third approach, we further loose the constraint of the input image to arbitrary in the wild images. Our proposed approach can robustly reconstruct high quality model even with extreme expressions and large poses. We then explore the applicability of our face reconstructions on four interesting applications: video face beautification, generating personalized facial blendshape from image sequences, face video stylizing and video face replacement. We demonstrate great potentials of our reconstruction approaches on these real-world applications. In particular, with the recent surge of interests in VR/AR, it is increasingly common to see people wearing head-mounted displays. However, the large occlusion on face is a big obstacle for people to communicate in a face-to-face manner. Our another application is that we explore hardware/software solutions for synthesizing the face image with presence of HMDs. We design two setups (experimental and mobile) which integrate two near IR cameras and one color camera to solve this problem. With our algorithm and prototype, we can achieve photo-realistic results. We further propose a deep neutral network to solve the HMD removal problem considering it as a face inpainting problem. This approach doesn't need special hardware and run in real-time with satisfying results.
5

AFFECT-PRESERVING VISUAL PRIVACY PROTECTION

Xu, Wanxin 01 January 2018 (has links)
The prevalence of wireless networks and the convenience of mobile cameras enable many new video applications other than security and entertainment. From behavioral diagnosis to wellness monitoring, cameras are increasing used for observations in various educational and medical settings. Videos collected for such applications are considered protected health information under privacy laws in many countries. Visual privacy protection techniques, such as blurring or object removal, can be used to mitigate privacy concern, but they also obliterate important visual cues of affect and social behaviors that are crucial for the target applications. In this dissertation, we propose to balance the privacy protection and the utility of the data by preserving the privacy-insensitive information, such as pose and expression, which is useful in many applications involving visual understanding. The Intellectual Merits of the dissertation include a novel framework for visual privacy protection by manipulating facial image and body shape of individuals, which: (1) is able to conceal the identity of individuals; (2) provide a way to preserve the utility of the data, such as expression and pose information; (3) balance the utility of the data and capacity of the privacy protection. The Broader Impacts of the dissertation focus on the significance of privacy protection on visual data, and the inadequacy of current privacy enhancing technologies in preserving affect and behavioral attributes of the visual content, which are highly useful for behavior observation in educational and medical settings. This work in this dissertation represents one of the first attempts in achieving both goals simultaneously.
6

Reconstrução de oclusões parciais em imagens de face visando o reconhecimento biométrico / Reconstruction of partial occlusions in face images for biometric recognition

Targino, Jonas Mendonça 10 December 2018 (has links)
Há um crescente incentivo ao uso da tecnologia biométrica para melhorar, e até mesmo substituir os métodos tradicionais de segurança. O campo da Biometria refere-se a uma grande variedade de tecnologias usadas para identicar ou vericar a identidade de uma pessoa por meio da mensuração e análise de vários aspectos físicos e comportamentais do ser humano. Modalidades biométricas são características extraídas do corpo humano, que são únicas para cada indivíduo e que podem ser usadas para estabelecer sua identidade numa população. As principais modalidades biométricas empregadas na literatura são: impressão digital, face, voz, palma da mão e íris. Dentre as modalidades biométricas, a face é a mais comumente vista e usada em nossa vida diária. Em aplicações de mundo real, sistemas de reconhecimento facial, frequentemente, têm que lidar com condições não controladas e não previsíveis, tais como mudança na iluminação, pose, expressão e oclusão, as quais introduzem variações intraclasse e degradam a performance do reconhecimento. Comparada com problemas de pose, iluminação e expressão, o problema relacionado à oclusão é relativamente pouco estudado na área. As abordagens empregadas para atacar o problema de reconhecimento facial na presença de oclusões parciais podem ser categorizadas em: i) abordagens baseadas em reconstrução, que formulam o reconhecimento biométrico baseado em imagens de face parcialmente ocluídas como um problema de reconstrução; ii) abordagens baseadas em correspondência local, que extraem características locais da face; iii) abordagens baseadas em características insensíveis, que extraem características da face que são robustas à oclusão. O objetivo desta dissertação é realizar um estudo comparativo detalhado das abordagens baseadas em reconstrução visando reconhecimento biométrico facial na presença de oclusão. As abordagens baseadas em reconstrução foram aplicadas a duas bases de dados (AR e Yale) contendo quatro tipos diferentes de oclusões e avaliadas usando a performance obtida no reconhecimento biométrico. Com base no estudo comparativo realizado, foram propostas três técnicas baseadas em reconstrução, que correspondem a uma modicação e/ou combinação de técnicas já conhecidas na literatura. Os resultados experimentais mostraram que as técnicas propostas alcançaram desempenhos melhores ou similares as técnicas descritas na literatura / There is a growing incentive to use biometric technology to improve, even replace, traditional security methods. The eld of Biometrics refers to a wide variety of technologies used to identify or verify a persons identity by measuring and analyzing various physical and behavioral aspects of the human being. Biometric modalities are characteristics drawn from the human body, which are unique to each individual and can be used to establish their identity in a population. The main biometric modalities used in the literature are: fingerprint, face, voice, palm, and iris. The main biometric modalities used in the literature are: fingerprint, face, voice, palm and iris. Among the biometric modalities, the face is the most commonly seen and used in our daily life. In real-world applications, facial recognition systems often have to deal with uncontrolled and unpredictable conditions such as change in illumination, pose, expression, and occlusion, which introduce intraclass variations and degrade recognition performance. Compared with problems of pose, illumination and expression, the problem related to occlusion is relatively little studied in the area. The approaches used to tackle facial recognition in the presence of partial occlusions can be categorized into: (i) reconstruction-based approaches that formulate recognition of occluded faces as a reconstruction problem; ii) approaches based on local correspondence, which extract local face characteristics; iii) approaches based on insensitive characteristics, which extract features of the face that are robust to occlusion. The objective of this dissertation is to perform a detailed comparative study of reconstruction-based approaches aiming at facial recognition in the presence of occlusion. Reconstruction-based approaches were applied to two databases (AR and Yale) containing four dierent types of occlusions and evaluated using the performance obtained in the biometric recognition. Based on the comparative study, three reconstruction-based techniques were proposed, which correspond to a change in the formulation and/or combination of techniques already known in the literature. The experimental results showed that the proposed techniques achieved better or similar performances than the techniques described in the literature
7

Using 3D morphable models for 3D photo-realistic personalized avatars and 2D face recognition / Les modèles déformables 3D (3DMM) pour des avatars personnalisables photo-réalistes et la reconnaissance de visages 2D

Zhou, Dianle 05 July 2011 (has links)
[Non communiqué] / In the past decade, 3D statistical face model (3D Morphable Model) has received much attention by both the commercial and public sectors. It can be used for face modeling for photo-realistic personalized 3D avatars and for the application 2D face recognition technique in biometrics. This thesis describes how to achieve an automatic 3D face reconstruction system that could be helpful for building photo-realistic personalized 3D avatars and for 2D face recognition with pose variability. The first systems we propose Combined Active Shape Model for 2D frontal facial landmark location and its application in 2D frontal face recognition in degraded condition. The second proposal is 3D Active Shape Model (3D-ASM) algorithm which is presented to automatically locate facial landmarks from different views. The third contribution is to use biometric data (2D images and 3D scan ground truth) for quantitatively evaluating the 3D face reconstruction. Finally, we address the issue of automatic 2D face recognition across pose using 3D Morphable Model
8

Vyhledávání korespondence ve stereosnímcích tváří / Stereoscopic Face Images Matching

Klaudíny, Martin Unknown Date (has links)
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.
9

Resurrection of Our Ancestors : 3D face reconstruction of historical images and its use in face recognition

Lekander, Moa Li January 2023 (has links)
In the field of face recognition one challenge is to recognize individuals in photographs taken of the person in profile. In this project it was investigated whether 3D face reconstructions could improve the face recognition results in these cases. Images used in the experiments were historical images of persons who lived in Stockholm during the late nineteenth to early twentieth centuries. Additionally, face frontalization algorithms were applied to images taken in profile-view of the person to investigate if this idea could improve the results of 3D face reconstruction of images in profile-view. The results of the experiments showed that using 3D face reconstructions did not improve the results of face recognition. However, in most cases the individuals in the 3D face reconstructions could be recognized. Furthermore, experiments also showed that face frontalization did not improve the results of the 3D face reconstructions of images taken in profile angle. Aside from these results, the thesis demonstrates a, to the best of our knowledge, new approach to evaluate 3D face reconstruction results by using face recognition.
10

Reconstruction multi-vues et texturation

Aganj, Ehsan 11 December 2009 (has links) (PDF)
Dans cette thèse, nous étudions les problèmes de reconstruction statique et dynamique à partir de vues multiples et texturation, en s'appuyant sur des applications réelles et pratiques. Nous proposons trois méthodes de reconstruction destinées à l'estimation d'une représentation d'une scène statique/dynamique à partir d'un ensemble d'images/vidéos. Nous considérons ensuite le problème de texturation multi-vues en se concentrant sur la qualité visuelle de rendu..

Page generated in 0.2503 seconds