Frédo Durand

Biography

Professor Fredo Durand received his PhD from Grenoble University (located in France) in 1999.Durand is a professor in the Electrical Engineering and Computer Science Department of the Massachusetts Institute of Technology, and a member of the Computer Science and Artificial Intelligence Laboratory (CSAIL). From 1999 to 2002, Durand was a post-doctoral researcher in the Computer Graphics Group where he is not an associate professor. He worked with Claude Puech and George Drettakis on both theoretical and practical aspects of 3D visibility. Some of his awards include: inaugural Eurographics Young Researcher Award (2004), an NSF CAREER award (2005), and Sloan fellowship (2006).

 

Industry Impact

Fredo works both on synthetic image generation and computational photography, where new algorithms afford powerful image enhancement and the design of imaging system that can record richer information about a scene. His research interests span most aspects of picture generation and creation, with emphasis on mathematical analysis, signal processing, and inspiration from perceptual sciences.

Research/Thesis Topic

Recent Works

Computational Understanding of Visualizations and Graphics Designs
This research focuses on using state-of-the-art deep learning approaches to automatically detect and parse text inside posters, to look around an image to automatically detect representative visual pictographs or icons, and to make predictions about the topics or concepts being communicated. The ability to make predictions about where people look on posters and graphs help to automatically generate text and visual summaries and thumbnails.

Sculpting by Numbers
This study proposes a method that allows an unskilled user to create an accurate physical replica of a digital 3D model. With the use of a projector/camera pair to scan a work in progress, and project multiple forms of guidance onto the object itself that indicates which areas need more material, which need less, and where any ridges, valleys or depth discontinuities are. We should how this approach can be used to create a duplicate of an existing object, by scanning the object and using that scan as the target shape. We demonstrate an end-to-end system in which real-world performance capture data is retargeted to Claymation. This approach allows users to easily and accurately create complex shapes, and naturally supports a large range of materials and model size.

Computational Bounce Flash for Indoor Portraits
Portraits taken with direct flash look harsh and unflattering because the light source comes from a small set of angles very close to the camera. Advanced photographers address this problem by using bounce flash, a technique where the flash is directed towards other surfaces in the room, creating a larger, virtual light source that can be cast from different directions to provide better shading variation for 3D modeling. Hence, to achieve high-quality photo lighting, a prototype camera dynamically reconstructs a 3D scene model and directs a motor-controlled flash head at nearby walls and ceilings for soft indirect illumination.