A5003065539- Face recognition and analysis
- Robotic Path Planning Algorithms
- Computational Geometry and Mesh Generation
- 3D Shape Modeling and Analysis
- Human Motion and Animation
- Generative Adversarial Networks and Image Synthesis
- Robotics and Sensor-Based Localization
- Computer Graphics and Visualization Techniques
- Facial Nerve Paralysis Treatment and Research
- Facial Rejuvenation and Surgery Techniques
- Advanced Vision and Imaging
- Modular Robots and Swarm Intelligence
- Autonomous Vehicle Technology and Safety
- Advanced Image and Video Retrieval Techniques
- Advanced Neural Network Applications
- Parallel Computing and Optimization Techniques
- Cardiac electrophysiology and arrhythmias
- Fluid Dynamics Simulations and Interactions
- Human Pose and Action Recognition
- Face Recognition and Perception
- 3D Modeling in Geospatial Applications
- Opportunistic and Delay-Tolerant Networks
- Anatomy and Medical Technology
- Advanced Image Processing Techniques
- Graph Theory and Algorithms
InterDigital (United States)
2022
Technicolor (Germany)
2018
Porsolt (France)
2017
Institut National des Sciences Appliquées de Rennes
2010-2014
Université de Rennes
2013
Université Européenne de Bretagne
2011
Institut de Recherche en Informatique et Systèmes Aléatoires
2011
Institut national de recherche en informatique et en automatique
2011
Abstract We present a general method for transferring skeletons and skinning weights between characters with distinct mesh topologies. Our pipeline takes as inputs source character rig (consisting of mesh, transformation hierarchy joints, weights) target mesh. From these inputs, we compute joint locations orientations that embed the skeleton in well to bind geometry new skeleton. consists two key steps. first geometric correspondence meshes using semi‐automatic relying on set markers. The...
In this paper we present a new technique to dynamically adapt the first step (broad phase) of collision detection process on hardware architecture during simulation. Our approach enables face unpredictable evolution simulation scenario (this includes addition complex objects, deletion, split into several ...). dynamic adaptation is performed sequential CPU, multi-core, single GPU and multi-GPU architectures. We propose use off-line simulations determine fields optimal performance for broad...
In this paper, we present a fully automatic pipeline for generating and stylizing high geometric textural quality facial rigs. They are automatically rigged with blendshapes animation, can be used across platforms applications including virtual reality, augmented remote collaboration, gaming more. From set of input photos, our approach is to able create photorealistic, character in less than seven minutes. The mesh reconstruction based on state-of-the art photogrammetry approaches. Automatic...
Abstract High‐fidelity digital human modeling has become crucial in various applications, including gaming, visual effects and virtual reality. Despite the significant impact of eyelashes on facial aesthetics, their reconstruction have been largely unexplored. In this paper, we introduce first data‐driven generative model based semantic features. This is derived from real data by introducing a new 3D eyelash method multi‐view images. The reconstructed made available which constitutes dataset...
Immersive media applications offer experiences that immerse the user in a virtual or hybrid environment and more degrees of freedom than with traditional 2D video content. Platforms immersive allow users to interact content other shared mixed-reality spaces. To address need for an interoperable cross-platform exchange format, interactive time-evolving solution such 3D environments, International Organization Standardization/International Electrotechnical Commission (ISO/IEC) Joint Technical...
While digital humans are key aspects of the rapidly evolving areas virtual reality, gaming, and online communications, many applications would benefit from using personalized (stylized) representations users, as they were shown to highly increase immersion, presence emotional response. In particular, depending on target application, one may want look like a dwarf or an elf in heroic fantasy world, alien another planet, accordance with style narrative. creating such replicas requires...
Facial caricature is the art of drawing faces in an exaggerated way to convey emotions such as humor or sarcasm. Automatic caricaturization has been explored both 2D and 3D domain. In this paper, we propose two novel approaches automatically caricaturize input facial scans, filling gaps literature terms user-control, style transfer, exploring use deep learning for mesh caricaturization. The first approach a gradient-based differential deformation with data driven stylization. It combination...
Abstract Even if the appearance and geometry of human eye have been extensively studied during last decade, geometrical correlation between gaze direction, eyelids aperture shape has not empirically modeled. In this paper, we propose a data‐driven approach for capturing modeling subtle features region, such as inner corner skin bulging effect due to globe orientation. Our consists an original experimental setup capture region variations combined with 3D reconstruction method. Regarding...
Our work is a perceptual study into the effects of training poses on Example-Based Facial Rigging (EBFR) method. We analyse output EBFR given set to see how well results reproduced our ground truth actor scans compared Deformation Transfer approach. While produced better overall, there were cases that did not any improvement. some these may be explained by lack sufficient for area face in question, we found certain lip improved training, despite large number mouth supplied.
In this paper, we propose a distributed and anticipative model for collision detection lead handling, two key components of physically-based simulations virtual environments. This is designed to improve the scalability interactive deterministic on systems such as PC clusters. Our main contribution consists loosening synchronism constraints in response pipeline allow simulation run decentralized, fashion. To do so, setup spatial subdivision grid, assign subset space each processor, made...
In this paper, we present FaceTuneGAN, a new 3D face model representation decomposing and encoding separately facial identity expression. We propose first adaptation of image-to-image translation networks, that have successfully been used in the 2D domain, to geometry. Leveraging recently released large scan databases, neural network has trained decouple factors variations with better knowledge face, enabling expressions transfer neutralization expressive faces. Specifically, design an...
To address the need for interoperable user representations (avatars) cross-platform exchange format immersive realities, ISO/IEC JTC 1/SC29/WG03 MPEG Systems has standardized a Scene Description framework in 23090-14 [ISO/IEC 2023]. It serves as baseline representation to enrich interactive experience between 3D objects an scene. This work presents Original Reference Geometric Avatar Neutral (Morgan), humanoid avatar specified informative content MPEG-I (MPEG-I SD) standardization group....