Next Billion Cameras

From Next Billion Cameras

(Redirected from Main Page)
Jump to: navigation, search

We encourage everyone to contribute to questions, add your own references and thoughts. Please click on 'Edit' link on the right for the corresponding section. Do NOT click on the top level Edit link.

Contents

Siggraph 2009 Course: Next Billion Cameras

What will be the impact of billions of available photos (with text and location information) on vision and graphics research and applications? Online image collections (e.g., Flickr, Google), already capture the significant tourist sites from almost every conceivable viewing position and angle, different times of day and night, and changes in season, weather, and decade. Furthermore, entire cities are now being captured from satellite, air, and street level. These collections are (or will soon be) likely dwarfed by the massive numbers of photos of people, whether they be celebrities or personal collections of family and friends. Events (e.g., the presidential inauguration) are similarly captured in both images and video. In short, this is the ultimate dataset---a dense sampling of the world's people, places, and things.

The capture and analysis of visual information plays an important role in photography, art, medical imaging, tele-presence, worker safety, scene understanding and robotics. But current computational approaches analyze images from single cameras that have only limited abilities. How will online photo collections transform visual social computing? What will a camera look like in ten years? How should we change the camera to improve mobile photography? How will a billion networked and portable cameras change the social culture? Cameras of the future will exploit unusual optics, novel illumination, and emerging sensors. A significant enhancement in the next billion cameras to support scene analysis, and mechanisms for superior metadata tagging for effective sharing will bring about a revolution in visual communication. We will explore the implications of these emerging devices and dataset for a range of research problems and applications in scene analysis and synthesis, including topics such as recognition, photo editing, 3D reconstruction, image-based rendering, photo editing, location estimation, and other areas.

A. Introduction‐‐5 minutes

B. Cameras of the future (Raskar, 30 minutes)

C. Reconstruction the World (Seitz, 30 minutes)

D. Understanding a Billion Photos (Efros, 30 minutes)

  • What will the photos depict?
  • Photos as visual content for computer graphics
  • Solving computer vision
  • PPT Slides


References

We encourage participants to add their own references here.

Help on Wiki

Retrieved from "http://scripts.mit.edu/~raskar/nextbillioncameras/index.php?title=Next_Billion_Cameras"
Personal tools