Return to SEMINARS

ERC Expressive seminar: Expressive cinematography (26/10/2015)

In the afternoon of Monday, 26 October 2015, Imagine organizes a seminar on Expressive cinematography. This event is a part of the series supported by the ERC Expressive grant. The seminar will be held in the Grand Amphi at Inria Grenoble, and is preceded by the PhD defense of Quentin Galvane in the morning (10:00) of the same day. Everyone is welcome!

Tentative Schedule

15:00 — 15:30
Philippe Guillotel (Technicolor)
Introduction to Haptic Cinematography

15:30 — 16:00
Arnav Jhala (University of California, Santa Cruz)
Computational Models of Narrative Semantics and their Connection to Expressive Cinematography

16:00 — 16:30
Mateu Sbert (University of Girona)
Non Photorealistic Rendering for Videogames and Cinema

16:30 — 17:00
Karan Singh (University of Toronto)
Cinematography in Mobile, Stereo, AR/VR and Evolving Formats


Philippe Guillotel : Introduction to Haptic Cinematography

Adding the sense of touch to hearing and viewing is necessary for a real immersive experience. This is the promise of the coming “4D-cinema” based on motion platforms and other sensory effects (water spray, wind, scent…). It is also going to be a mandatory additional cue for coming virtual reality applications, especially with full immersive head mounted displays. It provides a new dimension for filmmakers and leads to a new creative area, the haptic cinematography. In this talk we will describe the haptic cinematography workflow to create, distribute and render physical effects in addition to audio and visual cues. We will also address the creative space, since there is a clear need for some guiding rules to define a proper experience for the user, and not the current “waouh effects” which is generally not a long term solution. This introduction to haptic creation, suggests that artists should be more familiar with haptics, or even that this new creative space should be studied in schools of arts to prepare the first generation of haptographers.

Philippe Guillotel received the MS degree in electrical engineering and information theory from the University of Rennes, France, in 1986, the engineering degree in telecommunications, and computer science from the Ecole Nationale Supérieure des Télécommunications, France, in 1988 and the PhD degree from the University of Rennes, France, in 2012. He is a distinguished scientist at Technicolor R&I, France, where he is in charge of various image processing research topics and exploration of new experiences for the user. He has worked in the media and entertainment industry for more than 20 years. His main research interests include video compression, image processing, human machine interfaces, user experiences and entertainment applications.


Arnav Jhala : Computational Models of Narrative Semantics and their Connection to Expressive Cinematography

One of the purposes of cinematography is presentation or communication of an underlying content structure that is often represented in the form of a narrative. This process involves formulation of communicative goals, content selection, and selection of low-level visual features that satisfy these goals. Computational models of narrative identify these levels as plot, sjhuzet, and discourse. In this talk, I will present recent work on learning semantic graphical representations of narrative at the plot and sjhuzet level from text through crowdsourcing. Using a case-based reasoning algorithm called imaginative recall, I will demonstrate how principled transformations at the semantic level leads to textual narratives at different levels of complexity. Towards the end of the talk, I will present a proposal to initiate discussion on the applicability of similar approaches to expressive visual storytelling.

Arnav Jhala is an Assistant Professor of Computational Media at the University of California, Santa Cruz. His research interests lie at the intersection of artificial intelligence and digital media, particularly in the areas of computer games, cinematic communication, and narrative discourse. Arnav has developed computational models of film idioms and algorithms for automatically generating cinematic discourse. He heads the Computational Cinematics Studio at UC Santa Cruz.


Mateu Sbert : Non Photorealistic Rendering for Videogames and Cinema

Non Photorealistic Rendering (NPR) techniques are powerful expressive tools that allow to enrich a narrative. They can provide styles and moods, conveying different emotional and experiential representations of a graphical story. In this talk I will present our research on NPR screen based effects for videogames, and its posterior adaptation to 2D and 3D cinema.

Mateu Sbert is a full professor in Computer Science at the University of Girona, Spain. His research interests include the application of Monte Carlo and Information Theory techniques to Computer Graphics, Image Processing and Visualization. He has authored or co-authored more than 200 papers and three books in his areas of interest, participated in Eurographics, Siggraph Asia and Visweek tutorials, served as a member of program committee in many international conferences and as associate editor of CGF and IJCICG journals. He coordinated the GameTools project from the VIth European Framework.


Karan Singh : Cinematography in Mobile, Stereo, AR/VR and Evolving Formats

Cinematographic principles in use today are grounded in a century of film-making and at least a millennium of fine art. With the emergence of mobile, stereo, VR and AR devices of varying size, format and depth fidelity, there is now a need more than ever for tools to explore and evolve cinematographic vocabularies specific to these devices and environments. This talk will share a few experiences related to storyboard, editing and other authoring tools in the mobile, stereo and VR space.

Karan Singh is a Professor in Computer Science at the Univ. of Toronto. His research interests lie in artist driven interactive graphics, spanning geometric and anatomic modeling, character animation and sketch based interfaces. He has been a technical lead on two commercial projects that won technical Oscars (Maya, Paraform). He leads the design of two research systems based on sketch and sculpt metaphors (ILoveSketch, MeshMixer), that have been featured on leading design forums, and co-directs a reputed graphics and HCI lab, DGP.