Vous êtes ici

Séminaire du Jeudi 15/05/2014

Date: 
Jeudi, Mai 15, 2014 - 14:00 to 16:00
localisation: 
IMB, Bat A33, Salle de conférence

Orateur : Pierre Bénard

Affiliation : LaBRI - manao project

Titre : Stylizing Animation By Example

Résumé :Skilled artists, using traditional media or modern computer painting tools, can create a variety of expressive styles that are very appeal- ing in still images, but have been unsuitable for animation. The key difficulty is that existing techniques lack adequate temporal coherence to animate these styles effectively. Here we augment the range of practical animation styles by extending the guided texture syn- thesis method of Image Analogies [Hertzmann et al. 2001] to create temporally coherent animation sequences. To make the method art directable, we allow artists to paint portions of keyframes that are used as constraints. The in-betweens calculated by our method maintain stylistic continuity and yet change no more than necessary over time.

-------------------------------------------------------
Orateur : Hicham Badri

Affiliation : Inria Bordeaux Sud-Ouest, Equipe GeoStat

Titre : Fast Sparse Methods For Edge-Aware Processing

Résumé : Edge-Aware smoothing has received a lot of attention lately in the graphics community. It is the building block for a wide range of applications including detail manipulation, HDR tone-mapping, non-photorealistic rendering, sparse scribbles propagation, to cite a few.
Unfortunately, edge-aware filtering is a complex operation and robust methods are slow for real-time processing of images and videos. In this talk, we present new fast edge-aware operators based on effective sparse priors. We show how a first order approximation in the proximal form can help solving complex non-convex norms and thus allow using elaborated norms, leading to a better smoothing behavior. We show how to derive an efficient warm-start solution which leads to reduced processing time. We present various ways to perform faster processing using efficient preconditioning and parallel filtering. State-of-the art results are produced with reduced processing time.