Visual Worlds: Temporal Analysis, Animation and Authoring

News: We have a fully funded PhD offer to be started in fall 2023 on multi-valued implicit surface for complex shape modeling and animation - Details here
VISTA is a research group in Computer Graphics and Computer Vision at LIX laboratory, Ecole Polytechnique/CNRS, IP Paris.
From Real to Virtual, Visual Worlds, through the development of Analysis, Design and Animation of visual representations.

Team Keywords: Computer Graphics, Computer Vision with Deep Learning, Computer Animation, Expressive modeling, Visual Analysis, Video Understanding, Human Computer Interaction (HCI), Visual Computing, Real-Time Simulation, Virtual Worlds.

Short description

Visual representations are essential in many applications, from entertainment to science. VISTA is a Computer Graphics and Computer Vision group developing novel high level representations of visual and virtual contents for either fully automatic analysis and/or interactive real-time modification of 2D videos and 3D animated content.
On the one hand, we aim at proposing fully automatic understanding of existing visual material - including multiple modalities vastly available in videos-, extract spatio-temporal information, descriptive narrations from them and infer semantic concepts. On the second hand, we wish to place efficient new visual representations for 2D and 3D virtual models at the center of the interactive process with humans. These representations should be seen as a responsive, animation ready medium on which information is automatically extracted and analysed, while humans can freely interact, add elements and refine them in space and time according to their wish.

Detailed research description
More events

Recent Events

Were are participating to the French Workshop on AI & Video Games in presenting the ongoing work from David-Henri Garnier about bio-inspired modeling, and the participation of Marie-Paule Cani to the round table on the link between research and game companies.
We received the good news that the ANR proposal MultiForm will be funded. In collaboration with collegues from IRIT Toulouse (Loic Barthe) and Loria (Cedric Zanni), we will develop novel field-based representation for complex 3D model animation.
Robin Courant received the Best Student Paper Award Honorable Mention at ACCV 2022 for his work on FunnyNet: Audiovisual Learning of Funny moments in Videos done with Zhisong Liu (former PostDoc) and Vicky Kalogeiton. [tweet]
Jiayi Wei (VISTA and Geomerix) received the best paper award at the jFIG 2022 for her work on Robust-statistics Based Approach for Pointset Denoising. [tweet]

Research Axes

  • 1. Analysis and Understanding of Visual Content
    Deep CNN, Human-centric video learning
    Automatic & multimodal understanding
    Light learning, spatial representation
  • 2. Interactive Models for Shape and Motion
    Alternative representation (Field based, Implicit surfaces, ...)
    Spatio-temporal constraints
    Visual simulation, Layered models
    Behavioral simulation, Reinforcement learning
  • 3. Creating and Authoring Visual Worlds.
    Expressive creation: Sketching or Sculpting gestures, Sound, Multimodal system
    A-priori/learned knowledge constraints
    Narrative design, suggestion system
    Generation and style transfert, Visual transformers


  • - Interactive Visual models to Enhance Creativity
    Video Games, Metaverse, Animation Movies, VFX
    Shape Design, 3D Fabrication, Creative Arts
  • - Improve Entertainment, Safety and Social Good for Humans
    Movies, Serious Games, Cameras, Autonomous vehicles
  • - Support for Scientific Representation and Understanding
    Medical and Biological imaging, Archeology, Museography


Software & Code
Job Offers
Funded Projects