Rémi Ronfard


Researcher in Computer Vision and Machine Cinema
Member LEAR Team, INRIA Rhone-Alpes, Grenoble, France
INRIA Rhone-Alpes
655 Avenue de l'Europe
38330 Montbonnot
France
Phone:
Fax:
Office:
E-mail:
+33 4 7661 5303
H 116
remi.ronfard at inria dot fr

Publications

The complete list of my publications can now be found here.

News

The book Image and Geometry Processing for 3-D Cinematography, which I co-edited with Gabriel Taubin, has been published by Springer this summer.

You can buy it here.Check it out !









My habilitation thesis is now available in paperback !

It has been published by Editions Universitaires Europeennes this summer.

You can buy it here. Check it out !









Teaching

In January 2011, I will be teaching a course on Advanced 3D Animation to GMCAO Master 2 Pro students at University of Grenoble.

Masters internships

NEW ! Master 2 Internship Offers for 2010-2011 !

Prospective Master students should write to me (remi.ronfard at inria dot fr) for additional information.

PhD Thesis

I am offering the following topics for PhD theses: Prospective PhD students should write to me (remi.ronfard at inria dot fr) for additional information.

Research

I joined the LEAR team in April 2009 after a sabbatical in Montreal, where I was the head of the virtual cinematography team at Xtranormal Technologies, contributing to Xtranormal's "text-to-movie" technology.

My goal at Xtranormal was to provide automatic tools for placing cameras and lights in a virtual movie set, based on a narrative analysis of the movie script; and to automatically edit the shots into a complete movie, following a generic mathematical model of film grammar.

One my future research goals at INRIA is to extend such systems to a more varied combination of styles and contents. To that end, I am investigating new topics in the statistical analysis of film styles.

Using annotated movie databases, the goal of that research is to build statistical models of shot choices, camera framings, flighting and editing, with future applications in virtual movie making.

I am also interested in automatic video editing of live action footage. The additional difficulty with live footage is to describe each frame in terms of its content - where are the actors ? where are they looking ? what are they doing ? This is where computer vision meets film making.

I am also becoming interested in narrative analysis of video, i.e. understanding the causal chains of events and actions in an image sequence. For that purpose, I am revisiting my previous work in movie script alignement and analysis.

I am still investigating topics in visual action recognition. With my former Student Daniel Weinland and co-worker Edmond Boyer, I recently wrote a survey of the field, which has been accepted for publication in Computer Vision and image Understanding.

That survey paper is available on-line on the journal's site.

If you are a Master 2 student interested in any of the above topics, please check the following two offers for Master 2 Internships.

Habilitation

Nostalgia

Please visit my (older) personal home page including links to the Wizard of Oz demo and other related resources.