BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Seoul
X-LIC-LOCATION:Asia/Seoul
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:KST
DTSTART:18871231T000000
DTSTART:19881009T020000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20230103T035311Z
LOCATION:Room 323\, Level 3\, West Wing
DTSTART;TZID=Asia/Seoul:20221208T121500
DTEND;TZID=Asia/Seoul:20221208T123000
UID:siggraphasia_SIGGRAPH Asia 2022_sess258_gp_129@linklings.com
SUMMARY:Finding Nano-Ötzi: Cryo-Electron Tomography Visualization Guided b
 y Learned Segmentation
DESCRIPTION:Talks\n\nFinding Nano-Ötzi: Cryo-Electron Tomography Visualiza
 tion Guided by Learned Segmentation\n\nNguyen\n\nCryo-electron tomography 
 (cryo-ET) is a new 3D imaging technique with unprecedented potential for r
 esolving submicron structural details. Existing volume visualization metho
 ds, however, are not able to reveal details of interest due to low signal-
 to-noise ratio. In order to design more powerful transfer functions, we pr
 opose leveraging soft segmentation as an explicit component of visualizati
 on for noisy volumes. Our technical realization is based on semi-supervise
 d learning, where we combine the advantages of two segmentation algorithms
 . First, the weak segmentation algorithm provides good results for propaga
 ting sparse user-provided labels to other voxels in the same volume and is
  used to generate dense pseudo-labels. Second, the powerful deep-learning-
 based segmentation algorithm learns from these pseudo-labels to generalize
  the segmentation to other unseen volumes, a task that the weak segmentati
 on algorithm fails at completely. The proposed volume visualization uses d
 eep-learning-based segmentation as a component for segmentation-aware tran
 sfer function design. Appropriate ramp parameters can be suggested automat
 ically through frequency distribution analysis. Furthermore, our visualiza
 tion uses gradient-free ambient occlusion shading to further suppress the 
 visual presence of noise, and to give structural detail the desired promin
 ence. The cryo-ET data studied in our technical experiments are based on t
 he highest-quality tilted series of intact SARS-CoV-2 virions. Our techniq
 ue shows the high impact in target sciences for visual data analysis of ve
 ry noisy volumes that cannot be visualized with existing techniques.\n\nAu
 thors: Ngan Nguyen, Ciril Bohak, Dominik Engel, Peter Mindek, Ondřej Strna
 d, Peter Wonka, Sai Li, Timo Ropinski, Ivan Viola\n\nRegistration Category
 : FULL ACCESS, EXPERIENCE PLUS ACCESS, TRADE EXHIBITOR\n\nLanguage: ENGLIS
 H\n\nFormat: IN-PERSON
URL:https://sa2022.siggraph.org/en/full-program/?id=gp_129&sess=sess258
END:VEVENT
END:VCALENDAR