ログイン or 無料登録

Possibly novice photographers

日時 :
2021/04/30 (金) 13:30 ~
主催者 :

Possibly novice photographers and videographers who rely on their handheld devices to bite photos or make videos usually consider their subject’s lighting. Lighting is important in filmmaking, gaming, and virtual/augmented reality environments and may make or break the caliber of a scene and the actors and performers inside it. Replicating realistic character lighting possesses remained a difficult challenge in computer graphics and computer perspective.

While significant progress has already been made on volumetric capture methods, focusing on 3D geometric reconstruction by using high-resolution textures, such as an approach to achieve realistic shapes and textures of the human face, much less work has been done to recover photometric properties needed in view of relighting characters. Results from such systems lack fine details and also the subject’s shading is prebaked in to the texture.

Computer scientists at Google are revolutionizing this division of volumetric capture technology with a novel, comprehensive system that is able, for the first time, for you to capture full-body reflectance of 3D IMAGES human performances, and seamlessly blend them in to the real world through AR or even into digital scenes in videos, games, and more. Google will show their new system, called The Relightables, at ACM SIGGRAPH Indonesia, held Nov. 17 to 10 in Brisbane, Australia. SIGGRAPH Asia, now in its 12th calendar year, attracts the most respected complex and creative people from worldwide in computer graphics, animation, interactivity, video gaming, and emerging technologies.

There are actually major advances in this realm of work the industry calls 3D capture programs. Through these sophisticated systems, viewers have been able to experience digital characters become more active on the big screen, in particular, in blockbusters such as Avatar and the Avengers series and much more.

Indeed, the volumetric capture technology has reached a higher level of quality, but several reconstructions still lack true photorealism. In particular, despite these systems using high-end facilities setups with green screens, they still find it difficult to capture high-frequency details of humans and so they only recover a fixed illumination condition. This makes these volumetric catch systems unsuitable for photorealistic making of actors or performers around arbitrary scenes under different lights conditions.

Google’s Relightables system makes it possible to customize lighting on characters instantly or re-light them in any given scene or environment.

They demonstrate this on subjects which can be recorded inside a custom geodesic world outfitted with 331 custom colouring LED lights (also called your Light Stage capture system), several high-resolution cameras, and a couple of custom high-resolution depth sensors. The Relightables system captures concerning 65 GB per second involving raw data from nearly A HUNDRED cameras and its computational framework enables processing the info effectively at this scale. A video demonstration of the project is seen here:


  • このコメントは全員が閲覧、すべてのログインユーザーが投稿することができます