We capture the shape of moving cloth using a custom set of color markers printed on the surface of the cloth. The output is a sequence of triangle meshes with static connectivity and with detail at the scale of individual markers in both smooth and folded regions. We compute markers' coordinates in space using correspondence across multiple synchronized video cameras. Correspondence is determined from color information in small neighborhoods and refined using a novel strain pruning process. Final correspondence does not require neighborhood information. We use a novel data driven hole-filling technique to fill occluded regions. Our results include several challenging examples: a wrinkled shirt sleeve, a dancing pair of pants, and a rag tossed onto a cup. Finally, we demonstrate that cloth capture is reusable by animating a pair of pants using human motion capture data.
This video accompanies the publication "Capturing and Animating Occluded Cloth" by Ryan White, Keenan Crane, and David Forsyth in ACM SIGGRAPH 2007 (TOG).
For more information, see http://www.ryanmwhite.com/research/cloth_cap.html
Actually, Cinema 4D's Cloth module is able to do this exact same thing. But still, this is interesting technology.
niiidar 3 years ago
pretty awesome
shyninja 3 years ago
WOW! And I thought those are some funny pajamas! Reverse normal mapping.. The filling is very impressive too. After days of adjusting cloth sim parameters this seems like a dream. Would be interesting if you could do it in infrared leaving the opportunity to capture specular & diffuse data, possibly BDRF or Debevic's light probe sphere... endless possibilities, amazing work. -Shea
SheaDesign 3 years ago
amazing
caskachan 4 years ago