Siggraph Asia 2012 Paul Bourke Singapore Expo Centre Poster - - PowerPoint PPT Presentation

siggraph asia 2012
SMART_READER_LITE
LIVE PREVIEW

Siggraph Asia 2012 Paul Bourke Singapore Expo Centre Poster - - PowerPoint PPT Presentation

Siggraph Asia 2012 Paul Bourke Singapore Expo Centre Poster Poster Poster + emerging technology Poster + Emerging technology Emerging technology theme: touch Papers: Video Enhanced Gigapixel Panoramas Add localised motion to


slide-1
SLIDE 1

Siggraph Asia 2012

Paul Bourke

slide-2
SLIDE 2

Singapore Expo Centre

slide-3
SLIDE 3
slide-4
SLIDE 4

Poster

slide-5
SLIDE 5

Poster

slide-6
SLIDE 6
slide-7
SLIDE 7

Poster + emerging technology

slide-8
SLIDE 8
slide-9
SLIDE 9
slide-10
SLIDE 10

Poster + Emerging technology

slide-11
SLIDE 11
slide-12
SLIDE 12

Emerging technology “theme”: touch

slide-13
SLIDE 13
slide-14
SLIDE 14

Papers: Video Enhanced Gigapixel Panoramas

  • Add localised motion to gigapixel images.
  • Video region may move across the image.
  • Best of both worlds:
  • video has poor spatial resolution
  • gigapan images have poor temporal resolution.
  • Key to the work is the video registration/blending, and their interactive viewer.
slide-15
SLIDE 15
slide-16
SLIDE 16

Papers: Real or Fake? Human Judgments about Photographs and Computer-generated Images of Faces

  • Determine factors that cause people to to perceive images as real or CG.
  • Concentrated on faces
  • Outcomes
  • help cognitive scientists discover the mechanisms that underlie judgements of realism
  • help CG artist create more realistic characters
  • improve computational models in image forensics
  • “One key finding was that participants discriminated between photos and CG images best for
  • riginal images, less well for grayscale images, and worst for reflectance images. This suggests

that shading may be more important than colour for judgments about visual realism.“

  • Also experts outperformed non-expert but only for greyscale images.
slide-17
SLIDE 17
slide-18
SLIDE 18
slide-19
SLIDE 19

Paper: Data driven finger motion synthesis for gesturing characters

  • “In a controlled environment, we carefully captured and post-processed finger and body

motions from multiple actors. To augment the body motions of virtual characters with plausible and detailed finger movements, our method selects finger motion segments from the resulting database taking into account the similarity of the arm motions and the smoothness of consecutive finger motions.”

  • “Our key observation is that plausible finger motions can be inferred from the wrist motion.”
slide-20
SLIDE 20
slide-21
SLIDE 21

Art + emerging technology

slide-22
SLIDE 22

Siggraph Asia in Hong Kong in 2013

Tuesday November 19 - Friday November 22