BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250110T023312Z
LOCATION:Hall B5 (1)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241205T171600
DTEND;TZID=Asia/Tokyo:20241205T172800
UID:siggraphasia_SIGGRAPH Asia 2024_sess136_papers_104@linklings.com
SUMMARY:Learn to Create Simple LEGO Micro Buildings
DESCRIPTION:Technical Papers\n\nJiahao Ge, Mingjun Zhou, and Chi-Wing Fu (
 Chinese University of Hong Kong)\n\nThis paper presents the first learning
 -based generative pipeline for effectively creating 3D LEGO models. This t
 ask is very challenging due to the lack of dedicated representations and d
 atasets for learning coherently-connected bricks arrangements, as well as 
 an immense design space that is combinatorial in nature. We approach this 
 task by focusing on creating LEGO micro buildings. Our contributions are f
 our-fold. First, we propose the LEGO semantic volume representation to enc
 ode LEGO models, considering the bricks types and bricks connections, whil
 e allowing back-propagation learning. Second, we further consider the tran
 sformative nature of LEGO to atomize the semantic volume and formulate a g
 enerative model to learn the representation. Third, we build a rich datase
 t of LEGO micro buildings for model learning. Last, we design the progress
 ive LEGO decoder to reconstruct LEGO models from the generated representat
 ions, while ensuring bricks connections. We employed our pipeline to creat
 e LEGO micro buildings with a wide array of brick types, demonstrating its
  strong capability of learning diverse micro-building styles and producing
  assemble-able LEGO models. Further, we performed various quantitative eva
 luations, ablations, and a user study to show the compelling capability of
  our approach in terms of generative quality, fidelity, and diversity.\n\n
 Registration Category: Full Access, Full Access Supporter\n\nLanguage Form
 at: English Language\n\nSession Chair: Manolis Savva (Simon Fraser Univers
 ity)
URL:https://asia.siggraph.org/2024/program/?id=papers_104&sess=sess136
END:VEVENT
END:VCALENDAR